Back to index

4.11.0-0.ci-2024-06-06-035410

Jump to: Complete Features | Incomplete Features | Complete Epics | Incomplete Epics | Other Complete | Other Incomplete |

Changes from 4.10.67

Note: this page shows the Feature-Based Change Log for a release

Complete Features

These features were completed when this image was assembled

Problem:

Certain Insights Advisor features differentiate between RHEL and OCP advisor

Goal:

Address top priority UI misalignments between RHEL and OCP advisor. Address UI features dropped from Insights ADvisor for OCP GA.

 

Scope:

Specific tasks and priority of them tracked in https://issues.redhat.com/browse/CCXDEV-7432

 
 
 
 

 

This contains all the Insights Advisor widget deliverables for the OCP release 4.11.

Scope
It covers only minor bug fixes and improvements:

  • better error handling during internal outages in data processing
  • add "last refresh" timestamp in the Advisor widget
Scenario: Check if the Insights Advisor widget in the OCP WebConsole UI shows the time of the last data analysis
Given: OCP WebConsole UI and the cluster dashboard is accessible
And: CCX external data pipeline is in a working state
And: administrator A1 has access to his cluster's dashboard
And: Insights Operator for this cluster is sending archives
When: administrator A1 clicks on the Insights Advisor widget
Then: the results of the last analysis are showed in the Insights Advisor widget
And: the time of the last analysis is shown in the Insights Advisor widget 

Acceptance criteria:

  1. The time of the last analysis is shown in the Insights Advisor widget for the scenario above
  2. The way it is presented is defined within the scope of https://issues.redhat.com/browse/CCXDEV-5869 (mockup task)
  3. The source of this timestamp must be a result of running the Prometheus metric (last archive upload time):
    max_over_time(timestamp(changes(insightsclient_request_send_total\{status_code="202"}[1m]) > 0)[24h:1m])

Show the error message (mocked in CCXDEV-5868) if the Prometheus metrics `cluster_operator_conditions{name="insights"}` contain two true conditions: UploadDegraded and Degraded at the same time. This state occurs if there was an IO archive upload error = problems with the pipeline.

Expected for 4.11 OCP release.

The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Epic Goal

  • Allow admin user to create new alerting rules, targeting metrics in any namespace
  • Allow cloning of existing rules to simplify rule creation
  • Allow creation of silences for existing alert rules

Why is this important?

  • Currently, any platform-related metrics (exposed in a openshift-, kube- and default namespace) cannot be used to form a new alerting rule. That makes it very difficult for administrators to enrich our out of the box experience for the OpenShift Container Platform with new rules that may be specific to their environments.
  • Additionally, we had requests from customer to allow modifications of our existing, out of the box alerting rules (for instance tweaking the alert expression or changing the severity label). Unfortunately, that is not easy since most rules come from several open source projects, or other OpenShift components, and any modifications would make a seamless upgrade not really seamless anymore. Imagine K8s changes metrics again (see 1.14) and we have to update our rules. We would not know what modifications have been done (even just the threshold might be difficult if upstream changes that as well) and we would not be able to upgrade these rules.

Scenarios

  • I'd like to modify the query expression of an existing rule (because the threshold value doesn't match with my environment).

Cloning the existing rule should end up with a new rule in the same namespace.
Modifications can now be done to the new rule.
(Optional) You can silence the existing rule.

  • I'd like to create a new rule based on a metric only available to an openshift-* namespace

Create a new PrometheusRule object inside the namespace that includes the metrics you need to form the alerting rule.

  • I'd like to update the label of an existing rule.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • Ability to distinguish between rules deployed by us (CMO) and user created rules

Dependencies (internal and external)

Previous Work (Optional):

Open questions::

  1. Distinguish between operator-created rules and user-created rules
    Currently no such mechanism exists. This will need to be added to prometheus-operator or cluster-monitoring-operator.

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

CMO should reconcile the platform Prometheus configuration with the alert-relabel-config resources.

 

DoD

  • Alerts changed via alert-relabel-configs are evaluated by the Platform monitoring stack.
  • Product alerts which are overriden aren't sent to Alertmanager

CMO should reconcile the platform Prometheus configuration with the AlertingRule resources.

 

DoD

  • Alerts added via AlertingRule resources are evaluated by the Platform monitoring stack.

Managing PVs at scale for a fleet creates difficulties where "one size does not fit all". The ability for SRE to deploy prometheus with PVs and have retention based an on a desired size would enable easier management of these volumes across the fleet. 

 

The prometheus-operator exposes retentionSize.

Field Description
retentionSize Maximum amount of disk space used by blocks. Supported units: B, KB, MB, GB, TB, PB, EB. Ex: 512MB.

This is a feature request to enable this configuration option via CMO cluster-monitoring-config ConfigMap.

 

cc Simon Pasquier  

Epic Goal

  • Cluster admins want to configure the retention size for their metrics.

Why is this important?

  • While it is possible to define how long metrics should be retained on disk, it's not possible to tell the cluster monitoring operator how much data it should keep. For OSD/ROSA in particular, it would facilitate the management of the fleet if the retention size could be configured based on the persistent volume size because it would avoid issues with the storage getting full and monitoring being down when too many metrics are produced.

Scenarios

  • As a cluster admin, I want to define the maximum amount of data to be retained on the persistent volume.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • The cluster-monitoring-config config and the user-workload-monitoring-config configmap allow to configure the retention size for
    • Prometheus (Platform and UWM)
    • Thanos Ruler (to be confirmed)
  • Proper validation is in place preventing bad user inputs from breaking the stack.

Dependencies (internal and external)

  1. Thanos ruler doesn't support retention size (only retention time).

Previous Work (Optional):

  1. None

Open questions::

  1. None

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Problem Alignment

The Problem

Today, all configuration for setting individual, for example, routing configuration is done via a single configuration file that only admins have access to. If an environment uses multiple tenants and each tenant, for example, has different systems that they are using to notify teams in case of an issue, then someone needs to file a request w/ an admin to add the required settings.

That can be bothersome for individual teams, since requests like that usually disappear in the backlog of an administrator. At the same time, administrators might get tons of requests that they have to look at and prioritize, which takes them away from more crucial work.

We would like to introduce a more self service approach whereas individual teams can create their own configuration for their needs w/o the administrators involvement.

Last but not least, since Monitoring is deployed as a Core service of OpenShift there are multiple restrictions that the SRE team has to apply to all OSD and ROSA clusters. One restriction is the ability for customers to use the central Alertmanager that is owned and managed by the SRE team. They can't give access to the central managed secret due to security concerns so that users can add their own routing information.

High-Level Approach

Provide a new API (based on the Operator CRD approach) as part of the Prometheus Operator that allows creating a subset of the Alertmanager configuration without touching the central Alertmanager configuration file.

Please note that we do not plan to support additional individual webhooks with this work. Customers will need to deploy their own version of the third party webhooks.

Goal & Success

  • Allow users to deploy individual configurations that allow setting up Alertmanager for their needs without an administrator.

Solution Alignment

Key Capabilities

  • As an OpenShift administrator, I want to control who can CRUD individual configuration so that I can make sure that any unknown third person can touch the central Alertmanager instance shipped within OpenShift Monitoring.
  • As a team owner, I want to deploy a routing configuration to push notifications for alerts to my system of choice.

Key Flows

Team A wants to send all their important notifications to a specific Slack channel.

  • Administrator gives permission to Team A to allow creating a new configuration CR in their individual namespace.
  • Team A creates a new configuration CR.
  • Team A configures what alerts should go into their Slack channel.
  • Open Questions & Key Decisions (optional)
  • Do we want to improve anything inside the developer console to allow configuration?

Epic Goal

  • Allow users to manage Alertmanager for user-defined alerts and have the feature being fully supported.

Why is this important?

  • Users want to configure alert notifications without admin intervention.
  • The feature is currently Tech Preview, it should be generally available to benefit a bigger audience.

Scenarios

  1. As a cluster admin, I can deploy an Alertmanager service dedicated for user-defined alerts (e.g. separated from the existing  Alertmanager already used for platform alerts).
  2. As an application developer, I can silence alerts from the OCP console.
  3. As an application developer, I'm not allowed to configure invalid AlertmanagerConfig objects.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • The AlertmanagerConfig CRD is v1beta1
  • The validating webhook service checking AlertmanagerConfig resources is highly-available.

Dependencies (internal and external)

  1. Prometheus operator upstream should migrate the AlertmanagerConfig CRD from v1alpha1 to v1beta1
  2. Console enhancements likely to be involved (see below).

Previous Work (Optional):

  1. Part of the feature is available as Tech Preview (MON-880).

Open questions:

  1. Coordination with the console team to support the Alertmanager service dedicated for user-defined alerts.
  2. Migration steps for users that are already using the v1alpha1 CRD.

Done Checklist

 * CI - CI is running, tests are automated and merged.
 * Release Enablement <link to Feature Enablement Presentation>
 * DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
 * DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
 * DEV - Downstream build attached to advisory: <link to errata>
 * QE - Test plans in Polarion: <link or reference to Polarion>
 * QE - Automated tests merged: <link or reference to automated tests>
 * DOC - Downstream documentation merged: <link to meaningful PR> 

 

Now that upstream supports AlertmanagerConfig v1beta1 (see MON-2290 and https://github.com/prometheus-operator/prometheus-operator/pull/4709), it should be deployed by CMO.

DoD:

  • Kubernetes API exposes and supports the v1beta1 version for AlertmanagerConfig CRD (in addition to v1alpha1).
  • Users can manage AlertmanagerConfig v1beta1 objects seamlessly.
  • AlertmanagerConfig v1beta1 objects are reconciled in the generated Alertmanager configuration.
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Epic Goal

  • The goal is to support metrics federation for user-defined monitoring via the /federate Prometheus endpoint (both from within and outside of the cluster).

Why is this important?

  • It is already possible to configure remote write for user-defined monitoring to push metrics outside of the cluster but in some cases, the network flow can only go from the outside to the cluster and not the opposite. This makes it impossible to leverage remote write.
  • It is already possible to use the /federate endpoint for the platform Prometheus (via the internal service or via the OpenShift route) so not supporting for UWM doesn't provide a consistent experience.
  • If we don't expose the /federate endpoint for the UWM Prometheus, users would have no supported way to store and query application metrics from a central location.

Scenarios

  1. As a cluster admin, I want to federate user-defined metrics using the Prometheus /federate endpoint.
  2. As a cluster admin, I want that the /federate endpoint to UWM is accessible via an OpenShift route.
  3. As a cluster admin, I want that the access to the /federate endpoint to UWM requires authentication (with bearer token only) & authorization (the required permissions should match the permissions on the /federate endpoint of the Platform Prometheus).

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • Documentation - information about the recommendations and limitations/caveats of the federation approach.
  • User can federate user-defined metrics from within the cluster
  • User can federate user-defined metrics from the outside via the OpenShift route.

Dependencies (internal and external)

  1. None

Previous Work (Optional):

  1. None

Open questions:

  1. None

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

DoD

  • User can federate UWM metrics from outside of the cluster via the OpenShift route.
  • E2E test added to the CMO test suite.

DoD

  • User can federate UWM metrics within the cluster from the prometheus-user-workload.openshift-user-workload-monitoring.svc:9092 service
  • The service requires authentication via bearer token and authorization (same permissions as for federating platform metrics)

Copy/paste from [_https://github.com/openshift-cs/managed-openshift/issues/60_]

Which service is this feature request for?
OpenShift Dedicated and Red Hat OpenShift Service on AWS

What are you trying to do?
Allow ROSA/OSD to integrate with AWS Managed Prometheus.

Describe the solution you'd like
Remote-write of metrics is supported in OpenShift but it does not work with AWS Managed Prometheus since AWS Managed Prometheus requires AWS SigV4 auth.

  • Note that Prometheus supports AWS SigV4 since v2.26 and OpenShift 4.9 uses v2.29.

Describe alternatives you've considered
There is the workaround to use the "AWS SigV4 Proxy" but I'd think this is not properly supported by RH.
https://mobb.ninja/docs/rosa/cluster-metrics-to-aws-prometheus/

Additional context
The customer wants to use an open and portable solution to centralize metrics storage and analysis. If they also deploy to other clouds, they don't want to have to re-configure. Since most clouds offer a Prometheus service (or it's easy to self-manage Prometheus), app migration should be simplified.

Epic Goal

The cluster monitoring operator should allow OpenShift customers to configure remote write with all authentication methods supported by upstream Prometheus.

We will extend CMO's configuration API to support the following authentications with remote write:

  • Sigv4
  • Authorization
  • OAuth2

Why is this important?

Customers want to send metrics to AWS Managed Prometheus that require sigv4 authentication (see https://docs.aws.amazon.com/prometheus/latest/userguide/AMP-secure-metric-ingestion.html#AMP-secure-auth).

Scenarios

  1. As a cluster admin, I want to forward platform/user metrics to remote write systems requiring Sigv4 authentication.
  2. As a cluster admin, I want to forward platform/user metrics to remote write systems requiring OAuth2 authentication.
  3. As a cluster admin, I want to forward platform/user metrics to remote write systems requiring custom Authorization header for authentication (e.g. API key).

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • It is possible for a cluster admin to configure any authentication method that is supported by Prometheus upstream for remote write (both platform and user-defined metrics):
    • Sigv4
    • Authorization
    • OAuth2

Dependencies (internal and external)

  • In theory none because everything is already supported by the Prometheus operator upstream. We may discover bugs in the upstream implementation though that may require upstream involvement.

Previous Work

  • After CMO started exposing the RemoteWrite specification in MON-1069, additional authentication options where added to prometheus and prometheus-operator but CMO didn't catch up on these.

Open Questions

  • None

Prometheus and Prometheus operator already support sigv4 authentication for remote write. This should be possible to configure the same in the CMO configuration:

 

apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      remoteWrite:
      - url: "https://remote-write.endpoint"
        sigv4:
          accessKey:
            name: aws-credentialss
            key: access
          secretKey:
            name: aws-credentials
            key: secret

          profile: "SomeProfile"

          roleArn: "SomeRoleArn"

DoD:

  • Ability to configure sigv4 authentication for remote write in the openshift-monitoring/cluster-monitoring-config configmap
  • Ability to configure sigv4 authentication for remote write in the openshift-user-workload-monitoring/user-workload-monitoring-config configmap

Prometheus and Prometheus operator already support custom Authorization for remote write. This should be possible to configure the same in the CMO configuration:

 

apiVersion: v1
kind: ConfigMap
metadata:
  name: cluster-monitoring-config
  namespace: openshift-monitoring
data:
  config.yaml: |
    prometheusK8s:
      remoteWrite:
      - url: "https://remote-write.endpoint"
        Authorization:
          type: Bearer
          credentials:
            name: credentials
            key: token

DoD:

  • Ability to configure custom Authorization for remote write in the openshift-monitoring/cluster-monitoring-config configmap
  • Ability to configure custom Authorization for remote write in the openshift-user-workload-monitoring/user-workload-monitoring-config configmap
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Description

As WMCO user, I want to make sure containerd logging information has been updated in documents and scripts.

Acceptance Criteria

  • update must-gather to collect containerd logs
  • Internal/Customer Documents and log collecting scripts must have containerd specific information (ex: location of logs). 

Summary (PM+lead)

Configure audit logging to capture login, logout and login failure details

Motivation (PM+lead)

TODO(PM): update this

Customer who needs login, logout and login failure details inside the openshift container platform.
I have checked for this on my test cluster but the audit logs do not contain any user name specifying login or logout details. For successful logins or logout, on CLI and openshift console as well we can see 'Login successful' or 'Invalid credentials'.

Expected results: Login, logout and login failures should be captured in audit logging.

Goals (lead)

  1. Login, logout and login failures should be captured in audit logs

Non-Goals (lead)

  1. Don't attempt to log login failures in the IdP login flow that goes beyond timeout, if it the information is not available in explicit oauth-server requests (e.g. github password login error).
  2. Logout does not involve oauth-server (but is a simple API object deletion in oauth-apiserver). Hence, the audit log discussed here won't include logout.

Deliverables

  1. Changes to oauth-server to log into /varLog/oauth-server/audit.log on the master node.
  2. Documentation

Proposal (lead)

The apiserver pods today have ´/var/log/<kube|oauth|openshift>-apiserver` mounted from the host and create audit files there using the upstream audit event format (JSON lines following https://github.com/kubernetes/apiserver/blob/92392ef22153d75b3645b0ae339f89c12767fb52/pkg/apis/audit/v1/types.go#L72). These events are apiserver specific, but as oauth authentication flow events are also requests, we can use the apiserver event format to log logins, login failures and logouts. Hence, we propose to make oauth-server to create /var/log/oauth-server/audit.log files on the master nodes using that format.

When the login flow does not finish within a certain time (e.g. 10min), we can artificially create an event to show a login failure in the audit logs.

User Stories (PM)

Dependencies (internal and external, lead)

Previous Work (lead)

Open questions (lead)

  1. ...

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

 

🏆 What

Let the Cluster Authentication Operator deliver the policy to OAuthServer.

💖 Why

In order to know if authn events should be logged, OAuthServer needs to be aware of it.

🗒 Notes

Create an observer to deliver the audit policy to the oauth server

Make the authentication-operator react to the new audit field in the oauth.config/cluster object. Write an observer watching this field, such an observer will translate the top-level configuration into oauth-server config and add it to the rest of the observed config.

* Stanislav Láznička

OCP/Telco Definition of Done
Feature Template descriptions and documentation.

Feature Overview.

Early customer feedback is that they see SNO as a great solution covering smaller footprint deployment, but are wondering what is the evolution story OpenShift is going to provide where more capacity or high availability are needed in the future.

While migration tooling (moving workload/config to new cluster) could be a mid-term solution, customer desire is not to include extra hardware to be involved in this process.

 For Telecommunications Providers, at the Far Edge they intend to start small and then grow. Many of these operators will start with a SNO-based DU deployment as an initial investment, but as DUs evolve, different segments of the radio spectrum are added, various radio hardware is provisioned and features delivered to the Far Edge, the Telecommunication Providers desire the ability for their Far Edge deployments to scale up from 1 node to 2 nodes to n nodes. On the opposite side of the spectrum from SNO is MMIMO where there is a robust cluster and workloads use HPA.

Goals

  • Provide the capability to expand a single replica control plane topology to host more workloads capacity - add worker
  • Provide the capability to expand a single replica control plane to be a highly available control plane
  • To satisfy MMIMO Telecommunications providers will want the ability to scale a SNO to a multi-node cluster that can support HPA.
  • Telecommunications providers do not want workload (DU specifically) downtime when migrating from SNO to a multi-node cluster.
  • Telecommunications providers wish to be able to scale from one to two or more nodes to support a variety of radio hardware.
  • Support CP scaling (CP HA) for 2 node cluster, 3 node cluster and n node cluster. As the number of nodes in the cluster increases so does the failure domain of the cluster. The cluster is now supporting more cell sectors and therefore has more of a need for HA and resiliency including the cluster CP.

Requirements

  • TBD
Requirement Notes isMvp?
CI - MUST be running successfully with test automation This is a requirement for ALL features. YES
Release Technical Enablement Provide necessary release enablement details and documents. YES

(Optional) Use Cases

This Section:

  • Main success scenarios - high-level user stories
  • Alternate flow/scenarios - high-level user stories
  • ...

Questions to answer…

  • ...

Out of Scope

Background, and strategic fit

This Section: What does the person writing code, testing, documenting need to know? What context can be provided to frame this feature.

Assumptions

  • ...

Customer Considerations

  • ...

Documentation Considerations

Questions to be addressed:

  • What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
  • Does this feature have doc impact?
  • New Content, Updates to existing content, Release Note, or No Doc Impact
  • If unsure and no Technical Writer is available, please contact Content Strategy.
  • What concepts do customers need to understand to be successful in [action]?
  • How do we expect customers will use the feature? For what purpose(s)?
  • What reference material might a customer want/need to complete [action]?
  • Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
  • What is the doc impact (New Content, Updates to existing content, or Release Note)?

Epic Goal

  • Documented and supported flow for adding 1, 2, 3 or more workers to a Single Node OpenShift (SNO) deployment without requiring cluster downtime and the understanding that this action will not make the cluster itself highly available.

Why is this important?

  • Telecommunications and Edge scenarios where HA is handled via failover to another site but single site capacity may vary or need to be expanded over time.
  • Similar scenarios exist for some ISV vendors where OpenShift is an implementation detail of how they deliver their solution on top of another platform (e.g. VMware).

Scenarios

  1. Adding a worker to a single node openshift cluster.
  2. Adding a second worker to a single node openshift cluster.
  3. Adding a third worker to a single node openshift cluster.
  4. Removing a worker node from a single node openshift cluster that has had 1 or more workers added.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • Customer facing documentation of the add worker flow for SNO.

Dependencies (internal and external)

  1. ...

Previous Work (Optional):

Open questions::

  1. Presumably there is a scale limit on how many workers could be added to an SNO control plane, and it is lower than the limit for a "normal" 3 node control plane. It is not anticipated that this limit will be established in this epic. Intent is to focus on small scale sites where adding 1-3 worker nodes would be beneficial.

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

OCP/Telco Definition of Done
Epic Template descriptions and documentation.

<--- Cut-n-Paste the entire contents of this description into your new Epic --->

Epic Goal

  • Rebase OpenShift components to k8s v1.24

Why is this important?

  • Rebasing ensures components work with the upcoming release of Kubernetes
  • Address tech debt related to upstream deprecations and removals.

Scenarios

  1. ...

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • ...

Dependencies (internal and external)

  1. k8s 1.24 release

Previous Work (Optional):

Open questions::

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Feature Overview

  • As an infrastructure owner, I want a repeatable method to quickly deploy the initial OpenShift cluster.
  • As an infrastructure owner, I want to install the first (management, hub, “cluster 0”) cluster to manage other (standalone, hub, spoke, hub of hubs) clusters.

Goals

  • Enable customers and partners to successfully deploy a single “first” cluster in disconnected, on-premises settings

Requirements

4.11 MVP Requirements

  • Customers and partners needs to be able to download the installer
  • Enable customers and partners to deploy a single “first” cluster (cluster 0) using single node, compact, or highly available topologies in disconnected, on-premises settings
  • Installer must support advanced network settings such as static IP assignments, VLANs and NIC bonding for on-premises metal use cases, as well as DHCP and PXE provisioning environments.
  • Installer needs to support automation, including integration with third-party deployment tools, as well as user-driven deployments.
  • In the MVP automation has higher priority than interactive, user-driven deployments.
  • For bare metal deployments, we cannot assume that users will provide us the credentials to manage hosts via their BMCs.
  • Installer should prioritize support for platforms None, baremetal, and VMware.
  • The installer will focus on a single version of OpenShift, and a different build artifact will be produced for each different version.
  • The installer must not depend on a connected registry; however, the installer can optionally use a previously mirrored registry within the disconnected environment.

Use Cases

  • As a Telco partner engineer (Site Engineer, Specialist, Field Engineer), I want to deploy an OpenShift cluster in production with limited or no additional hardware and don’t intend to deploy more OpenShift clusters [Isolated edge experience].
  • As a Enterprise infrastructure owner, I want to manage the lifecycle of multiple clusters in 1 or more sites by first installing the first  (management, hub, “cluster 0”) cluster to manage other (standalone, hub, spoke, hub of hubs) clusters [Cluster before your cluster].
  • As a Partner, I want to package OpenShift for large scale and/or distributed topology with my own software and/or hardware solution.
  • As a large enterprise customer or Service Provider, I want to install a “HyperShift Tugboat” OpenShift cluster in order to offer a hosted OpenShift control plane at scale to my consumers (DevOps Engineers, tenants) that allows for fleet-level provisioning for low CAPEX and OPEX, much like AKS or GKE [Hypershift].
  • As a new, novice to intermediate user (Enterprise Admin/Consumer, Telco Partner integrator, RH Solution Architect), I want to quickly deploy a small OpenShift cluster for Poc/Demo/Research purposes.

Questions to answer…

  •  

Out of Scope

Out of scope use cases (that are part of the Kubeframe/factory project):

  • As a Partner (OEMs, ISVs), I want to install and pre-configure OpenShift with my hardware/software in my disconnected factory, while allowing further (minimal) reconfiguration of a subset of capabilities later at a different site by different set of users (end customer) [Embedded OpenShift].
  • As an Infrastructure Admin at an Enterprise customer with multiple remote sites, I want to pre-provision OpenShift centrally prior to shipping and activating the clusters in remote sites.

Background, and strategic fit

  • This Section: What does the person writing code, testing, documenting need to know? What context can be provided to frame this feature.

Assumptions

  1. The user has only access to the target nodes that will form the cluster and will boot them with the image presented locally via a USB stick. This scenario is common in sites with restricted access such as government infra where only users with security clearance can interact with the installation, where software is allowed to enter in the premises (in a USB, DVD, SD card, etc.) but never allowed to come back out. Users can't enter supporting devices such as laptops or phones.
  2. The user has access to the target nodes remotely to their BMCs (e.g. iDrac, iLo) and can map an image as virtual media from their computer. This scenario is common in data centers where the customer provides network access to the BMCs of the target nodes.
  3. We cannot assume that we will have access to a computer to run an installer or installer helper software.

Customer Considerations

  • ...

Documentation Considerations

Questions to be addressed:

  • What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
  • Does this feature have doc impact?
  • New Content, Updates to existing content, Release Note, or No Doc Impact
  • If unsure and no Technical Writer is available, please contact Content Strategy.
  • What concepts do customers need to understand to be successful in [action]?
  • How do we expect customers will use the feature? For what purpose(s)?
  • What reference material might a customer want/need to complete [action]?
  • Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
  • What is the doc impact (New Content, Updates to existing content, or Release Note)?

 

References

 

 

Epic Goal

  • As an OpenShift infrastructure owner, I need to be able to integrate the installation of my first on-premises OpenShift cluster with my automation flows and tools.
  • As an OpenShift infrastructure owner, I must be able to provide the CLI tool with manifests that contain the definition of the cluster I want to deploy
  • As an OpenShift Infrastructure owner, I must be able to get the validation errors in a programmatic way
  • As an OpenShift Infrastructure owner, I must be able to get the events and progress of the installation in a programmatic way
  • As an OpenShift Infrastructure owner, I must be able to retrieve the kubeconfig and OpenShift Console URL in a programmatic way

Why is this important?

  • When deploying clusters with a large number of hosts and when deploying many clusters, it is common to require to automate the installations.
  • Customers and partners usually use third party tools of their own to orchestrate the installation.
  • For Telco RAN deployments, Telco partners need to repeatably deploy multiple OpenShift clusters in parallel to multiple sites at-scale, with no human intervention.

Scenarios

  1. Monitoring flow:
    1. I generate all the manifests for the cluster,
    2. call the CLI tool pointint to the manifests path,
    3. Obtain the installation image from the nodes
    4. Use my infrastructure capabilities to boot the image on the target nodes
    5. Use the tool to connect to assisted service to get validation status and events
    6. Use the tool to retrieve credentials and URL for the deployed cluster

Acceptance Criteria

  • Backward compatibility between OCP releases with automation manifests (they can be applied to a newer version of OCP).
  • Installation progress and events can be tracked programatically
  • Validation errors can be obtained programatically
  • Kubeconfig and console URL can be obtained programatically
  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

References

User Story:

As a deployer, I want to be able to:

  • Get the credentials for the cluster that is going to be deployed

so that I can achieve

  • Checking the installed cluster for installation completion
  • Connect and administer the cluster that gets installed

 

Currently the Assisted Service generates the credentials by running the ignition generation step of the oepnshift-installer. This is why the credentials are only retrievable from the REST API towards the end of the installation.

In the BILLI usage, which takes down assisted service before the installation is complete there is no obvious point at which to alert the user that they should retrieve the credentials. This means that we either need to:

  • Allow the user to pass the admin key that will then get signed by the generated CA and replace the key that is made by openshift-installer (would mean new functionality in AI)
  • Allow the key to be retrieved by SSH with the fleeting command from the node0 (after it has generated). The command should be able to wait until it is possible
  • Have the possibility to POST it somewhere

Acceptance Criteria:

  • The admin key is generated and usable to check for installation completeness

This requires/does not require a design proposal.
This requires/does not require a feature gate.

Feature Overview

The AWS-specific code added in OCPPLAN-6006 needs to become GA and with this we want to introduce a couple of Day2 improvements.
Currently the AWS tags are defined and applied at installation time only and saved in the infrastructure CRD's status field for further operator use, which in turn just add the tags during creation.

Saving in the status field means it's not included in Velero backups, which is a crucial feature for customers and Day2.
Thus the status.resourceTags field should be deprecated in favour of a newly created spec.resourceTags with the same content. The installer should only populate the spec, consumers of the infrastructure CRD must favour the spec over the status definition if both are supplied, otherwise the status should be honored and a warning shall be issued.

Being part of the spec, the behaviour should also tag existing resources that do not have the tags yet and once the tags in the infrastructure CRD are changed all the AWS resources should be updated accordingly.

On AWS this can be done without re-creating any resources (the behaviour is basically an upsert by tag key) and is possible without service interruption as it is a metadata operation.

Tag deletes continue to be out of scope, as the customer can still have custom tags applied to the resources that we do not want to delete.

Due to the ongoing intree/out of tree split on the cloud and CSI providers, this should not apply to clusters with intree providers (!= "external").

Once confident we have all components updated, we should introduce an end2end test that makes sure we never create resources that are untagged.

After that, we can remove the experimental flag and make this a GA feature.

Goals

  • Inclusion in the cluster backups
  • Flexibility of changing tags during cluster lifetime, without recreating the whole cluster

Requirements

  • This Section:* A list of specific needs or objectives that a Feature must deliver to satisfy the Feature.. Some requirements will be flagged as MVP. If an MVP gets shifted, the feature shifts. If a non MVP requirement slips, it does not shift the feature.
Requirement Notes isMvp?
CI - MUST be running successfully with test automation This is a requirement for ALL features. YES
Release Technical Enablement Provide necessary release enablement details and documents. YES

List any affected packages or components.

  • Installer
  • Cluster Infrastructure
  • Storage
  • Node
  • NetworkEdge
  • Internal Registry
  • CCO

RFE-1101 described user defined tags for AWS resources provisioned by an OCP cluster. Currently user can define tags which are added to the resources during creation. These tags cannot be updated subsequently. The propagation of the tags is controlled using experimental flag. Before this feature goes GA we should define and implement a mechanism to exclude any experimental flags. Day2 operations and deletion of tags is not in the scope.

RFE-2012 aims to make the user-defined resource tags feature GA. This means that user defined tags should be updatable.

Currently the user-defined tags during install are passed directly as parameters of the Machine and Machineset resources for the master and worker. As a result these tags cannot be updated by consulting the Infrastructure resource of the cluster where the user defined tags are written.

The MCO should be changed such that during provisioning the MCO looks up the values of the tags in the Infrastructure resource and adds the tags during creation of the EC2 resources. The MCO should also watch the infrastructure resource for changes and when the resource tags are updated it should update the tags on the EC2 instances without restarts.

Acceptance Criteria:

  • e2e test where the ResourceTags are updated and then the test verifies that the tags on the ec2 instances are updated without restarts. now moved to CFE-179

Feature Overview  

Much like core OpenShift operators, a standardized flow exists for OLM-managed operators to interact with the cluster in a specific way to leverage AWS STS authorization when using AWS APIs as opposed to insecure static, long-lived credentials. OLM-managed operators can implement integration with the CloudCredentialOperator in well-defined way to support this flow.

Goals:

Enable customers to easily leverage OpenShift's capabilities around AWS STS with layered products, for increased security posture. Enable OLM-managed operators to implement support for this in well-defined pattern.

Requirements:

  • CCO gets a new mode in which it can reconcile STS credential request for OLM-managed operators
  • A standardized flow is leveraged to guide users in discovering and preparing their AWS IAM policies and roles with permissions that are required for OLM-managed operators 
  • A standardized flow is defined in which users can configure OLM-managed operators to leverage AWS STS
  • An example operator is used to demonstrate the end2end functionality
  • Clear instructions and documentation for operator development teams to implement the required interaction with the CloudCredentialOperator to support this flow

Use Cases:

See Operators & STS slide deck.

 

Out of Scope:

  • handling OLM-managed operator updates in which AWS IAM permission requirements might change from one version to another (which requires user awareness and intervention)

 

Background:

The CloudCredentialsOperator already provides a powerful API for OpenShift's cluster core operator to request credentials and acquire them via short-lived tokens. This capability should be expanded to OLM-managed operators, specifically to Red Hat layered products that interact with AWS APIs. The process today is cumbersome to none-existent based on the operator in question and seen as an adoption blocker of OpenShift on AWS.

 

Customer Considerations

This is particularly important for ROSA customers. Customers are expected to be asked to pre-create the required IAM roles outside of OpenShift, which is deemed acceptable.

Documentation Considerations

  • Internal documentation needs to exists to guide Red Hat operator developer teams on the requirements and proposed implementation of integration with CCO and the proposed flow
  • External documentation needs to exist to guide users on:
    • how to become aware that the cluster is in STS mode
    • how to become aware of operators that support STS and the proposed CCO flow
    • how to become aware of the IAM permissions requirements of these operators
    • how to configure an operator in the proposed flow to interact with CCO

Interoperability Considerations

  • this needs to work with ROSA
  • this needs to work with self-managed OCP on AWS

Market Problem

This Section: High-Level description of the Market Problem ie: Executive Summary

  • As a customer of OpenShift layered products, I need to be able to fluidly, reliably and consistently install and use OpenShift layered product Kubernetes Operators into my ROSA STS clusters, while keeping a STS workflow throughout.
  •  
  • As a customer of OpenShift on the big cloud providers, overall I expect OpenShift as a platform to function equally well with tokenized cloud auth as it does with "mint-mode" IAM credentials. I expect the same from the Kubernetes Operators under the Red Hat brand (that need to reach cloud APIs) in that tokenized workflows are equally integrated and workable as with "mint-mode" IAM credentials.
  •  
  • As the managed services, including Hypershift teams, offering a downstream opinionated, supported and managed lifecycle of OpenShift (in the forms of ROSA, ARO, OSD on GCP, Hypershift, etc), the OpenShift platform should have as close as possible, native integration with core platform operators when clusters use tokenized cloud auth, driving the use of layered products.
  • .
  • As the Hypershift team, where the only credential mode for clusters/customers is STS (on AWS) , the Red Hat branded Operators that must reach the AWS API, should be enabled to work with STS credentials in a consistent, and automated fashion that allows customer to use those operators as easily as possible, driving the use of layered products.

Why it Matters

  • Adding consistent, automated layered product integrations to OpenShift would provide great added value to OpenShift as a platform, and its downstream offerings in Managed Cloud Services and related offerings.
  • Enabling Kuberenetes Operators (at first, Red Hat ones) on OpenShift for the "big3" cloud providers is a key differentiation and security requirement that our customers have been and continue to demand.
  • HyperShift is an STS-only architecture, which means that if our layered offerings via Operators cannot easily work with STS, then it would be blocking us from our broad product adoption goals.

Illustrative User Stories or Scenarios

  1. Main success scenario - high-level user story
    1. customer creates a ROSA STS or Hypershift cluster (AWS)
    2. customer wants basic (table-stakes) features such as AWS EFS or RHODS or Logging
    3. customer sees necessary tasks for preparing for the operator in OperatorHub from their cluster
    4. customer prepares AWS IAM/STS roles/policies in anticipation of the Operator they want, using what they get from OperatorHub
    5. customer's provides a very minimal set of parameters (AWS ARN of role(s) with policy) to the Operator's OperatorHub page
    6. The cluster can automatically setup the Operator, using the provided tokenized credentials and the Operator functions as expected
    7. Cluster and Operator upgrades are taken into account and automated
    8. The above steps 1-7 should apply similarly for Google Cloud and Microsoft Azure Cloud, with their respective token-based workload identity systems.
  2. Alternate flow/scenarios - high-level user stories
    1. The same as above, but the ROSA CLI would assist with AWS role/policy management
    2. The same as above, but the oc CLI would assist with cloud role/policy management (per respective cloud provider for the cluster)
  3. ...

Expected Outcomes

This Section: Articulates and defines the value proposition from a users point of view

  • See SDE-1868 as an example of what is needed, including design proposed, for current-day ROSA STS and by extension Hypershift.
  • Further research is required to accomodate the AWS STS equivalent systems of GCP and Azure
  • Order of priority at this time is
    • 1. AWS STS for ROSA and ROSA via HyperShift
    • 2. Microsoft Azure for ARO
    • 3. Google Cloud for OpenShift Dedicated on GCP

Effect

This Section: Effect is the expected outcome within the market. There are two dimensions of outcomes; growth or retention. This represents part of the “why” statement for a feature.

  • Growth is the acquisition of net new usage of the platform. This can be new workloads not previously able to be supported, new markets not previously considered, or new end users not previously served.
  • Retention is maintaining and expanding existing use of the platform. This can be more effective use of tools, competitive pressures, and ease of use improvements.
  • Both of growth and retention are the effect of this effort.
    • Customers have strict requirements around using only token-based cloud credential systems for workloads in their cloud accounts, which include OpenShift clusters in all forms.
      • We gain new customers from both those that have waited for token-based auth/auth from OpenShift and from those that are new to OpenShift, with strict requirements around cloud account access
      • We retain customers that are going thru both cloud-native and hybrid-cloud journeys that all inevitably see security requirements driving them towards token-based auth/auth.
      •  

References

As an engineer I want the capability to implement CI test cases that run at different intervals, be it daily, weekly so as to ensure downstream operators that are dependent on certain capabilities are not negatively impacted if changes in systems CCO interacts with change behavior.

Acceptance Criteria:

Create a stubbed out e2e test path in CCO and matching e2e calling code in release such that there exists a path to tests that verify working in an AWS STS workflow.

Feature Overview

Customers are asking for improvements to the upgrade experience (both over-the-air and disconnected). This is a feature tracking epics required to get that work done.  

Goals

  1. Have an option to do upgrades in more discrete steps under admin control. Specifically, these steps are: 
    • Control plane upgrade
    • Worker nodes upgrade
    • Workload enabling upgrade (i..e. Router, other components) or infra nodes
  2. Better visibility into any errors during the upgrades and documentation of what they error means and how to recover. 
  3. An user experience around an end-2-end back-up and restore after a failed upgrade 
  4. OTA-810  - Better Documentation: 
    • Backup procedures before upgrades. 
    • More control over worker upgrades (with tagged pools between user Vs admin)
    • The kinds of pre-upgrade tests that are run, the errors that are flagged and what they mean and how to address them. 
    • Better explanation of each discrete step in upgrades, and what each CVO Operator is doing and potential errors, troubleshooting and mitigating actions.

References

OCP/Telco Definition of Done
Epic Template descriptions and documentation.

<--- Cut-n-Paste the entire contents of this description into your new Epic --->

Epic Goal

  • Provide a one click option to perform an upgrade which pauses all non master pools

Why is this important?

  • Customers are increasingly asking that the overall upgrade is broken up into more digestible pieces
  • This is the limit of what's possible today
    • R&D work will be done in the future to allow for further bucketing of upgrades into Control Plane, Worker Nodes, and Workload Enabling components (ie: router) That will however take much more consideration and rearchitecting

Scenarios

  1. An admin selecting their upgrade is offered two options "Upgrade Cluster" and "Upgrade Control Plane"
    1. If the admin selects Upgrade Cluster they get the pre 4.10 behavior
    2. If the admin selects Upgrade Control Plane all non master pools are paused and an upgrade is initiated
  1. A tooltip should clarify what the difference between the two are
  2. The pool progress bars should indicate pause/unpaused status, non master pools should allow for unpausing

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • ...

Dependencies (internal and external)

  1. ...

Previous Work (Optional):

  1. While this epic doesn't specifically target upgrading from 4.N to 4.N+1 to 4.N+2 with non master pools paused it would fundamentally enable that and it would simplify the UX described in Paused Worker Pool Upgrades

Open questions::

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Goal
Add the ability to choose between a full cluster upgrade (which exists today) or control plane upgrade (which will pause all worker pools) in the console.

Background
Currently in the console, users only have the ability to complete a full cluster upgrade. For many customers, upgrades take longer than what their maintenance window allows. Users need the ability to upgrade the control plane independently of the other worker nodes. 

Ex. Upgrades of huge clusters may take too long so admins may do the control plane this weekend, worker-pool-A next weekend, worker-pool-B the weekend after, etc.  It is all at a pool level, they will not be able to choose specific hosts.

Requirements

  1. Changes to the Update modal:
    1. Add the ability to choose between a cluster upgrade and a control plane upgrade (the design does not default to a selection but rather disables the update button to force the user to make a conscious decision)
    2. link out to documentation to learn more about update strategies
  2. Changes to the in progress check list:
    1. Add a status above the worker pool section to let users know that all worker pools are paused and an action to resume all updates
    2. Add a "resume update" button for each worker pool entry
  3. Changes to the update status:
    1. When all master pools are updated successfully, change the status from what we have today "Up to date" to something like "Control plane up to date - all worker pools paused"
  4. Add an inline alert that lets users know there is a 60 day window to update all worker pools. In the alert, include the sentiment that worker pools can remain paused as long as is normally safe, which means until certificate rotation becomes critical which is at about 60 days. The admin would be advised to unpause them in order to complete the full upgrade. If the MCPs are paused, the certification rotation does not happen, which causes the cluster to become degraded and causes failure in multiple 'oc' commands, including but not limited to 'oc debug', 'oc logs', 'oc exec' and 'oc attach'. (Are we missing anything else here?) Inline alert logic:
    1. From day 60 to day 10 use the default alert.
    2. From day 10 to day 3 use the warning alert.
    3. From day 3 to 0 use the critical alert and continue to persist until resolved.

Design deliverables: 

Goal
Improve the UX on the machine config pool page to reflect the new enhancements on the cluster settings that allows users to select the ability to update the control plane only.

Background
Currently in the console, users only have the ability to complete a full cluster upgrade. For many customers, upgrades take longer than what their maintenance window allows. Users need the ability to upgrade the control plane independently of the other worker nodes. 

Ex. Upgrades of huge clusters may take too long so admins may do the control plane this weekend, worker-pool-A next weekend, worker-pool-B the weekend after, etc.  It is all at a pool level, they will not be able to choose specific hosts.

Requirements

  1. Changes to the table:
    1. Remove "Updated, updating and paused" columns. We could also consider adding column management to this table and hide those columns by default.
    2. Add "Update status" as a column, and surface the same status on cluster settings. Not true or false values but instead updating, paused, and up to date.
    3. Surface the update action in the table row.
  2. Add an inline alert that lets users know there is a 60 day window to update all worker pools. In the alert, include the sentiment that worker pools can remain paused as long as is normally safe, which means until certificate rotation becomes critical which is at about 60 days. The admin would be advised to unpause them in order to complete the full upgrade. If the MCPs are paused, the certification rotation does not happen, which causes the cluster to become degraded and causes failure in multiple 'oc' commands, including but not limited to 'oc debug', 'oc logs', 'oc exec' and 'oc attach'. (Are we missing anything else here?) Add the same alert logic to this page as the cluster settings:
    1. From day 60 to day 10 use the default inline alert.
    2. From day 10 to day 3 use the warning inline alert.
    3. From day 3 to 0 use the critical alert and continue to persist until resolved.

Design deliverables: 

Goal:
As a cluster administrator, I want OpenShift to include a recent HAProxy version, so that I have the latest available performance and security fixes.  

 Description:
We should strive to follow upstream HAProxy releases by bumping the HAProxy version that we ship in OpenShift with every 4.y release, so that OpenShift benefits from upstream performance and security fixes, and so that we avoid large version-number jumps when an urgent fix necessitates bumping to the latest HAProxy release.  This bump should happen as early as possible in the OpenShift release cycle, so as to maximize soak time.   

For OpenShift 4.13, this means bumping to 2.6.  

As a cluster administrator, 

I want OpenShift to include a recent HAProxy version, 

so that I have the latest available performance and security fixes.  

 

We should strive to follow upstream HAProxy releases by bumping the HAProxy version that we ship in OpenShift with every 4.y release, so that OpenShift benefits from upstream performance and security fixes, and so that we avoid large version-number jumps when an urgent fix necessitates bumping to the latest HAProxy release.  This bump should happen as early as possible in the OpenShift release cycle, so as to maximize soak time.   

For OpenShift 4.14, this means bumping to 2.6.  

Bump the HAProxy version in dist-git so that OCP 4.13 ships HAProxy 2.6.13, with this patch added on top: https://git.haproxy.org/?p=haproxy-2.6.git;a=commit;h=2b0aafdc92f691bc4b987300c9001a7cc3fb8d08. The patch fixes the segfault that was being tracked as OCPBUGS-13232.

This patch is in HAProxy 2.6.14, so we can stop carrying the patch once we bump to HAProxy 2.6.14 or newer in a subsequent OCP release.

OCP/Telco Definition of Done
Feature Template descriptions and documentation.
<--- Cut-n-Paste the entire contents of this description into your new Feature --->
<--- Remove the descriptive text as appropriate --->

Feature Overview

  • As RH OpenShift Product Owners, we want to enable new providers/platforms/service with varying levels of capabilities and integration with minimal reliance on OpenShift Engineering.
  • As a new provider/platform partner, I want to enable my solution (hardware and/or software) with OpenShift with minimal effort.

 

Problem

  • It is currently challenging for us to enable new platforms / providers without taking the heavy burden on doing the platform specific development ourselves.

Goals

  • We want to enable the long-tail new platforms/providers to expand our reach into new markets and/or support new use cases.
  • We want to remove strict dependencies we have on Engineering teams to review, support and test new providers.
  • We want to lower the effort required for onboarding new platforms/providers.
  • We want to enable new platform/providers to self-certify.
  • We want to define tiered model for provider/platform integration that delineates ownership and responsibilities throughout new provider/platform development lifecycle and support model.
  • We want to reduce time to onboard new provider/platform – ideally to a single release.
  • We want to maintain consistent customer experience across all providers/platforms.

Requirements

  • Step-by-step guide on how to add a new platform/provider for each tier
  • Certification tool for partner to self-certify
  • Certification tool results for (at least) each Y/minor release submitted by partner to Red Hat for acknowledgement
  • DCI program to enable partners to run CI with OpenShift on their platform
  • Well documented, accessible, and up-to-date test suites for providing the test coverage of the partner
  • CI includes upgrade testing of OpenShift with partner's components
  • Partner component upgrade failure should not block OpenShift upgrade
  • Partner code is available in repositories in the openshift org on github with an open source license compatible with OpenShift

 

Requirement Notes isMvp?
CI - MUST be running successfully with test automation This is a requirement for ALL features. YES
Release Technical Enablement Provide necessary release enablement details and documents. YES

(Optional) Use Cases

This Section:

  • Main success scenarios - high-level user stories
  • Alternate flow/scenarios - high-level user stories
  • ...

Questions to answer…

  • ...

Out of Scope

Background, and strategic fit

This Section: What does the person writing code, testing, documenting need to know? What context can be provided to frame this feature.

Assumptions

  • ...

Customer Considerations

  • ...

Documentation Considerations

Questions to be addressed:

  • What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
  • Does this feature have doc impact?
  • New Content, Updates to existing content, Release Note, or No Doc Impact
  • If unsure and no Technical Writer is available, please contact Content Strategy.
  • What concepts do customers need to understand to be successful in [action]?
  • How do we expect customers will use the feature? For what purpose(s)?
  • What reference material might a customer want/need to complete [action]?
  • Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
  • What is the doc impact (New Content, Updates to existing content, or Release Note)?

 

References

The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Running the OPCT with the latest version (v0.1.0) on OCP 4.11.0, the openshift-tests is reporting an incorrect counter for the "total" field.

In the example below, after the 1127th test, the total follows the same counter of executed. I also would assume that the total is incorrect before that point as the test continues the execution increases both counters.

 

openshift-tests output format: [failed/executed/total]

started: (0/1126/1127) "[sig-storage] PersistentVolumes-expansion  loopback local block volume should support online expansion on node [Suite:openshift/conformance/parallel] [Suite:k8s]"

passed: (38s) 2022-08-09T17:12:21 "[sig-storage] In-tree Volumes [Driver: nfs] [Testpattern: Dynamic PV (default fs)] provisioning should provision storage with mount options [Suite:openshift/conformance/parallel] [Suite:k8s]"

started: (0/1127/1127) "[sig-storage] In-tree Volumes [Driver: local][LocalVolumeType: tmpfs] [Testpattern: Generic Ephemeral-volume (block volmode) (late-binding)] ephemeral should support two pods which have the same volume definition [Suite:openshift/conformance/parallel] [Suite:k8s]"

passed: (6.6s) 2022-08-09T17:12:21 "[sig-storage] Downward API volume should provide container's memory request [NodeConformance] [Conformance] [Suite:openshift/conformance/parallel/minimal] [Suite:k8s]"

started: (0/1128/1128) "[sig-storage] In-tree Volumes [Driver: cinder] [Testpattern: Dynamic PV (immediate binding)] topology should fail to schedule a pod which has topologies that conflict with AllowedTopologies [Suite:openshift/conformance/parallel] [Suite:k8s]"

skip [k8s.io/kubernetes@v1.24.0/test/e2e/storage/framework/testsuite.go:116]: Driver local doesn't support GenericEphemeralVolume -- skipping
Ginkgo exit error 3: exit with code 3

skipped: (400ms) 2022-08-09T17:12:21 "[sig-storage] In-tree Volumes [Driver: local][LocalVolumeType: tmpfs] [Testpattern: Generic Ephemeral-volume (block volmode) (late-binding)] ephemeral should support two pods which have the same volume definition [Suite:openshift/conformance/parallel] [Suite:k8s]"

started: (0/1129/1129) "[sig-storage] In-tree Volumes [Driver: emptydir] [Testpattern: Dynamic PV (default fs)] capacity provides storage capacity information [Suite:openshift/conformance/parallel] [Suite:k8s]" 

 

OPCT output format [executed/total (failed failures)]

Tue, 09 Aug 2022 14:12:13 -03> Global Status: running
JOB_NAME                         | STATUS     | RESULTS    | PROGRESS                  | MESSAGE                                           
openshift-conformance-validated  | running    |            | 1112/1127 (0 failures)    | status=running                                    
openshift-kube-conformance       | complete   |            | 352/352 (0 failures)      | waiting for post-processor...                     
Tue, 09 Aug 2022 14:12:23 -03> Global Status: running
JOB_NAME                         | STATUS     | RESULTS    | PROGRESS                  | MESSAGE                                           
openshift-conformance-validated  | running    |            | 1120/1127 (0 failures)    | status=running                                    
openshift-kube-conformance       | complete   |            | 352/352 (0 failures)      | waiting for post-processor...                     
Tue, 09 Aug 2022 14:12:33 -03> Global Status: running
JOB_NAME                         | STATUS     | RESULTS    | PROGRESS                  | MESSAGE                                           
openshift-conformance-validated  | running    |            | 1139/1139 (0 failures)    | status=running                                    
openshift-kube-conformance       | complete   |            | 352/352 (0 failures)      | waiting for post-processor...                     
Tue, 09 Aug 2022 14:12:43 -03> Global Status: running
JOB_NAME                         | STATUS     | RESULTS    | PROGRESS                  | MESSAGE                                           
openshift-conformance-validated  | running    |            | 1185/1185 (0 failures)    | status=running                                    
openshift-kube-conformance       | complete   |            | 352/352 (0 failures)      | waiting for post-processor...                     
Tue, 09 Aug 2022 14:12:53 -03> Global Status: running
JOB_NAME                         | STATUS     | RESULTS    | PROGRESS                  | MESSAGE                                           
openshift-conformance-validated  | running    |            | 1188/1188 (0 failures)    | status=running                                    
openshift-kube-conformance       | complete   |            | 352/352 (0 failures)      | waiting for post-processor...      

 

 

 

 

Feature Overview (aka. Goal Summary)  

Customers can trust the metadata in our operators catalogs to reason about infrastructure compatibility and interoperability. Similar to OCPPLAN-7983 the requirement is that this data is present for every layered product and Red Hat-release operator and ideally also ISV operators.

Today it is hard to validate the presence of this data due to the metadata format. This features tracks introducing a new format, implementing the appropriate validation and enforcement of presence as well as defining a grace period in which both formats are acceptable.

Goals (aka. expected user outcomes)

Customers can rely on the operator metadata as the single source of truth for capability and interoperability information instead of having to look up product-specific documentation. They can use this data to filter in on-cluster and public catalog displays as well as in their pipelines or custom workflows.

Red Hat Operators are required to provide this data and we aim for near 100% coverage in our catalogs.

Absence of this data can reliably be detected and will subsequently lead to gating in the release process.

Requirements (aka. Acceptance Criteria):

  • discrete annotations per feature that can be checked for presence as well as positive and negative values (see PORTEANBLE-525)
  • support in the OCP console and RHEC to support both the new and the older metadata annotations format
  • enforcement in ISV and RHT operator release pipelines
    • first with non-fatal warnings
    • later with blocking behavior if annotations are missing
    • the presence of ALL annotations needs to be checked in all pipelines / catalogs

Questions to Answer:

  • when can we rollout the pipeline tests?
    • only when there is support for visualization in the OCP Console and catalog.redhat.com
  • should operator authors use both, old and new annotations at the same time?
    • they can, but there is no requirement to do that, once the support in console and RHEC is there, the pipelines will only check for the new annotations
  • what happens to older OCP releases that don't support the new annotations yet?
    • the only piece in OCP that is aware of the annotations is the console, and we plan to backport the changes all the way to 4.10

 

Customer Considerations

Provide any additional customer-specific considerations that must be made when designing and delivering the Feature.  Initial completion during Refinement status.

 

Documentation Considerations

  • we first need internal documentation for RHT and ISV teams that need to implement the change
  • when RHEC and Console are ready, we will update the external documentation and and can point to that as the official source of truth

 

Interoperability Considerations

  • OCP Console will have to support the new format (see CONSOLE-3688) in parallel to the old format (as fallback) in all currently supported OCP versions

Epic Goal

  • Transparently support old and new infrastructure annotations format delivered by OLM-packaged operators

Why is this important?

  • As part of part of OCPSTRAT-288 we are looking to improve the metadata quality of Red Hat operators in OpenShift
  • via PORTENABLE-525 we are defining a new metadata format that supports the aforementioned initiative with more robust detection of individual infrastructure features via boolean data types

Scenarios

  1. A user can use the OCP console to browse through the OperatorHub catalog and filter for all the existing and new annotations defined in PORTENABLE-525
  2. A user reviewing an operator's detail can see the supported infrastructures transparently regardless if the operator uses the new or the existing annotations format

Acceptance Criteria

  • the new annotation format is supported in operatorhub filtering and operator details pages
  • the old annotation format keeps being supported in operatorhub filtering and operator details pages
  • the console will respect both the old and the new annotations format
  • when for a particular feature both the operator denotes data in both the old and new annotation format, the annotations in the newer format take precedence
  • the newer infrastructure features from PORTENABLE-525 tls-profiles and token-auth/* do not have equivalents in the old annotation format and evaluation doesn't need to fall back as described in the previous point

Dependencies (internal and external)

  1. none

Open Questions

  1. due to the non-intrusive nature of this feature, can we ship it in a 4.14.z patch release?

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Goal

Increase integration of Shipwright, Tekton, Argo CD in OpenShift GitOps with OpenShift platform and related products such as ACM.

Incomplete Features

When this image was assembled, these features were not yet completed. Therefore, only the Jira Cards included here are part of this release

Feature Overview

We drive OpenShift cross-market customer success and new customer adoption with constant improvements and feature additions to the existing capabilities of our OpenShift Core Networking (SDN and Network Edge). This feature captures that natural progression of the product.

Goals

  • Feature enhancements (performance, scale, configuration, UX, ...)
  • Modernization (incorporation and productization of new technologies)

Requirements

  • Core Networking Stability
  • Core Networking Performance and Scale
  • Core Neworking Extensibility (Multus CNIs)
  • Core Networking UX (Observability)
  • Core Networking Security and Compliance

In Scope

  • Network Edge (ingress, DNS, LB)
  • SDN (CNI plugins, openshift-sdn, OVN, network policy, egressIP, egress Router, ...)
  • Networking Observability

Out of Scope

There are definitely grey areas, but in general:

  • CNV
  • Service Mesh
  • CNF

Documentation Considerations

Questions to be addressed:

  • What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
  • Does this feature have doc impact?
  • New Content, Updates to existing content, Release Note, or No Doc Impact
  • If unsure and no Technical Writer is available, please contact Content Strategy.
  • What concepts do customers need to understand to be successful in [action]?
  • How do we expect customers will use the feature? For what purpose(s)?
  • What reference material might a customer want/need to complete [action]?
  • Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
  • What is the doc impact (New Content, Updates to existing content, or Release Note)?
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

User Story: As a customer in a highly regulated environment, I need the ability to secure DNS traffic when forwarding requests to upstream resolvers so that I can ensure additional DNS traffic and data privacy.

The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Create a PR in openshift/cluster-ingress-operator to implement configurable router probe timeouts.

The PR should include the following:

  • Changes to the ingress operator's ingress controller to allow the user to configure the readiness and liveness probe's timeoutSeconds values.
  • Changes to existing unit tests to verify that the new functionality works properly.
  • Write E2E test to verify that the new functionality works properly.

tldr: three basic claims, the rest is explanation and one example

  1. We cannot improve long term maintainability solely by fixing bugs.
  2. Teams should be asked to produce designs for improving maintainability/debugability.
  3. Specific maintenance items (or investigation of maintenance items), should be placed into planning as peer to PM requests and explicitly prioritized against them.

While bugs are an important metric, fixing bugs is different than investing in maintainability and debugability. Investing in fixing bugs will help alleviate immediate problems, but doesn't improve the ability to address future problems. You (may) get a code base with fewer bugs, but when you add a new feature, it will still be hard to debug problems and interactions. This pushes a code base towards stagnation where it gets harder and harder to add features.

One alternative is to ask teams to produce ideas for how they would improve future maintainability and debugability instead of focusing on immediate bugs. This would produce designs that make problem determination, bug resolution, and future feature additions faster over time.

I have a concrete example of one such outcome of focusing on bugs vs quality. We have resolved many bugs about communication failures with ingress by finding problems with point-to-point network communication. We have fixed the individual bugs, but have not improved the code for future debugging. In so doing, we chase many hard to diagnose problem across the stack. The alternative is to create a point-to-point network connectivity capability. this would immediately improve bug resolution and stability (detection) for kuryr, ovs, legacy sdn, network-edge, kube-apiserver, openshift-apiserver, authentication, and console. Bug fixing does not produce the same impact.

We need more investment in our future selves. Saying, "teams should reserve this" doesn't seem to be universally effective. Perhaps an approach that directly asks for designs and impacts and then follows up by placing the items directly in planning and prioritizing against PM feature requests would give teams the confidence to invest in these areas and give broad exposure to systemic problems.


Relevant links:

In OCP 4.8 the router was changed to use the "random" balancing algorithm for non-passthrough routes by default. It was previously "leastconn".

Bug https://bugzilla.redhat.com/show_bug.cgi?id=2007581 shows that using "random" by default incurs significant memory overhead for each backend that uses it.

PR https://github.com/openshift/cluster-ingress-operator/pull/663
reverted the change and made "leastconn" the default again (OCP 4.8 onwards).

The analysis in https://bugzilla.redhat.com/show_bug.cgi?id=2007581#c40 shows that the default haproxy behaviour is to multiply the weight (specified in the route CR) by 16 as it builds its data structures for each backend. If no weight is specified then openshift-router sets the weight to 256. If you have many, many thousands of routes then this balloons quickly and leads to a significant increase in memory usage, as highlighted by customer cases attached to BZ#2007581.

The purpose of this issue is to both explore changing the openshift-router default weight (i.e., 256) to something smaller, or indeed unset (assuming no explicit weight has been requested), and to measure the memory usage within the context of the existing perf&scale tests that we use for vetting new haproxy releases.

It may be that the low-hanging change is to not default to weight=256 for backends that only have one pod replica (i.e., if no value specified, and there is only 1 pod replica, then don't default to 256 for that single server entry).

Outcome: does changing the [default] weight value make it feasible to switch back to "random" as the default balancing algorithm for a future OCP release.

Revert router to using "random" once again in 4.11 once analysis is done on impact of weight and static memory allocation.

Per the 4.6.30 Monitoring DNS Post Mortem, we should add E2E tests to openshift/cluster-dns-operator to reduce the risk that changes to our CoreDNS configuration break DNS resolution for clients.  

To begin with, we add E2E DNS testing for 2 or 3 client libraries to establish a framework for testing DNS resolvers; the work of adding additional client libraries to this framework can be left for follow-up stories.  Two common libraries are Go's resolver and glibc's resolver.  A somewhat common library that is known to have quirks is musl libc's resolver, which uses a shorter timeout value than glibc's resolver and reportedly has issues with the EDNS0 protocol extension.  It would also make sense to test Java or other popular languages or runtimes that have their own resolvers. 

Additionally, as talked about in our DNS Issue Retro & Testing Coverage meeting on Feb 28th 2024, we also decided to add a test for testing a non-EDNS0 query for a larger than 512 byte record, as once was an issue in bug OCPBUGS-27397.   

The ultimate goal is that the test will inform us when a change to OpenShift's DNS or networking has an effect that may impact end-user applications. 

Feature Overview

  • This Section:* High-Level description of the feature ie: Executive Summary
  • Note: A Feature is a capability or a well defined set of functionality that delivers business value. Features can include additions or changes to existing functionality. Features can easily span multiple teams, and multiple releases.

 

Goals

  • This Section:* Provide high-level goal statement, providing user context and expected user outcome(s) for this feature

 

Requirements

  • This Section:* A list of specific needs or objectives that a Feature must deliver to satisfy the Feature.. Some requirements will be flagged as MVP. If an MVP gets shifted, the feature shifts. If a non MVP requirement slips, it does not shift the feature.

 

Requirement Notes isMvp?
CI - MUST be running successfully with test automation This is a requirement for ALL features. YES
Release Technical Enablement Provide necessary release enablement details and documents. YES

 

(Optional) Use Cases

This Section: 

  • Main success scenarios - high-level user stories
  • Alternate flow/scenarios - high-level user stories
  • ...

 

Questions to answer…

  • ...

 

Out of Scope

 

Background, and strategic fit

This Section: What does the person writing code, testing, documenting need to know? What context can be provided to frame this feature.

 

Assumptions

  • ...

 

Customer Considerations

  • ...

 

Documentation Considerations

Questions to be addressed:

  • What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
  • Does this feature have doc impact?  
  • New Content, Updates to existing content,  Release Note, or No Doc Impact
  • If unsure and no Technical Writer is available, please contact Content Strategy.
  • What concepts do customers need to understand to be successful in [action]?
  • How do we expect customers will use the feature? For what purpose(s)?
  • What reference material might a customer want/need to complete [action]?
  • Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
  • What is the doc impact (New Content, Updates to existing content, or Release Note)?
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

When viewing the Installed Operators list set to 'All projects' and then selecting an operator that is available in 'All namespaces' (globally installed,) upon clicking the operator to view its details the user is taken into the details of that operator in installed namespace (project selector will switch to the install namespace.)

This can be disorienting then to look at the lists of custom resource instances and see them all blank, since the lists are showing instances only in the currently selected project (the install namespace) and not across all namespaces the operator is available in.

It is likely that making use of the new Operator resource will improve this experience (CONSOLE-2240,) though that may still be some releases away. it should be considered if it's worth a "short term" fix in the meantime.

Note: The informational alert was not implemented. It was decided that since "All namespaces" is displayed in the radio button, the alert was not needed.

Goal
Add support for PDB (Pod Disruption Budget) to the console.

Requirements:

  • Add a list, detail, and yaml view (with samples) for PDBs. In addition, update the workloads page to support PDBs as well.
  • For the PBD list page include a table with name, namespace, selector, availability, allowed disruptions and created. In addition, to the table provide the main call to action to create a PDB.
  • For the PDB details page provide a Details, YAML and Pods tab. The Pods tab will include a list pods associated with the PBD - make sure to surface the owner column.
  • When users create a PDB from the list page, take them to the YAML and provide samples to enhance the creation experience. Sample 1: Set max unavailable to 0, Sample 2: Set min unavailable to 25% (confirming samples with stakeholders). In the case that a PDB has already been applied, warn users that it is not recommended to add another. Cover use cases as well that keep users from creating poor policies - for example, setting the minimum available to zero.
  • Add the ability to add/edit/view PBDs on a workload. If we edit a PDB applied to multiple workloads, warn users that this change will affect all workloads and not only the one they are currently editing. When a PDB has been applied, add a new filed to the details page with a link to the PDB and policy.

Designs:

Samuel Padgett Colleen Hart

During master nodes upgrade when nodes are getting drained there's currently no protection from two or more operands going down. If your component is required to be available during upgrade or other voluntary disruptions, please consider deploying PDB to protect your operands.

The effort is tracked in https://issues.redhat.com/browse/WRKLDS-293.

Example:

 

Acceptance Criteria:
1. Create PDB controller in console-operator for both console and downloads pods
2. Add e2e tests for PDB in single node and multi node cluster

 

Note: We should consider to backport this to 4.10

Feature Overview

Enable sharing ConfigMap and Secret across namespaces

Requirements

Requirement Notes isMvp?
Secrets and ConfigMaps can get shared across namespaces   YES

Questions to answer…

NA

Out of Scope

NA

Background, and strategic fit

Consumption of RHEL entitlements has been a challenge on OCP 4 since it moved to a cluster-based entitlement model compared to the node-based (RHEL subscription manager) entitlement mode. In order to provide a sufficiently similar experience to OCP 3, the entitlement certificates that are made available on the cluster (OCPBU-93) should be shared across namespaces in order to prevent the need for cluster admin to copy these entitlements in each namespace which leads to additional operational challenges for updating and refreshing them. 

Documentation Considerations

Questions to be addressed:
 * What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)?
 * Does this feature have doc impact?
 * New Content, Updates to existing content, Release Note, or No Doc Impact
 * If unsure and no Technical Writer is available, please contact Content Strategy.
 * What concepts do customers need to understand to be successful in [action]?
 * How do we expect customers will use the feature? For what purpose(s)?
 * What reference material might a customer want/need to complete [action]?
 * Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available.
 * What is the doc impact (New Content, Updates to existing content, or Release Note)?

OCP/Telco Definition of Done
Epic Template descriptions and documentation.

<--- Cut-n-Paste the entire contents of this description into your new Epic --->

Epic Goal

  • Require volumes that use the Shared Resources CSI driver to specify readOnly: true in order to create the pod
  • Reserve the "openshift-" prefix for SharedSecrets and SharedConfigMaps, such that these resources can only be created by OpenShift operators. We must do this while the driver is tech preview.

Why is this important?

  • readOnly: true must be specified in order for the driver to mount the volume correctly. If this is not set, the volume mount is rejected and the pod will be stuck in a Pending/Initializing state.
  • A validating admission webhook will ensure that the pods won't be created in such a state, improving user experience.
  • Openshift operators may want/need to create SharedSecrets and SharedConfigMaps so they can be used as system level resources. For example, Insights Operator can automatically create a SharedSecret for the Simple Content Access cert.

Scenarios

  1. As a developer, I want to consume shared Secrets and ConfigMaps in my workloads so that I can have access to shared credentials and configuration.
  2. As a cluster admin, I want the Insights operator to automatically create a SharedSecret for my cluster's simple content access certificate.
  3. As a cluster admin/SRE, I want OpenShift to use SharedConfigMaps to distribute cluster certificate authorities so that data is not duplicated in ConfigMaps across my cluster.

Acceptance Criteria

  • Pods must have readOnly: true set to use the shared resource CSI Driver - admission should be rejected if this is not set.
  • Documentation updated to reflect this requirement.
  • Users (admins?) are not allowed to create SharedSecrets or SharedConfigMaps with the "openshift-" prefix.

Dependencies (internal and external)

  1. ART - to create payload image for the webhook
  2. Arch review for the enhancement proposal (Apiserver/control plane team)

Previous Work (Optional):

  1. BUILD-293 - Shared Resources tech preview

Open questions::

  1. From email exchange with David Eads:  "Thinking ahead to how we'd like to use this in builds once we're GA, are we likely to choose openshift-etc-pki-entitlement as one of our well-known names?  If we do, what sort of validation (if any) would we like to provide on the backing secret and does that require any new infrastructure?"

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

User Story

As a developer using SharedSecrets and ConfigMaps
I want to ensure all pods set readOnly; true on admission
So that I don't have pods stuck in the "Pending" state because of a bad volume mount

Acceptance Criteria

  • Pods which reference the Shared Resource CSI driver must set readOnly: true on admission.
  • If readOnly: true is not set, or is set to false, the pod should not be created.
  • Appropriate testing in place to verify behavior

QE Impact

QE will need to verify the new Pod Admission behavior

Docs Impact

Docs will need to ensure that readOnly: true is required and must be set to true.

PX Impact

None.

QE testing/verification of the feature - require readOnly to be true

Actions:

1. Create smoke test and submit to GitHub
2. Run script to integrate smoke test with Polarion

User Story

As an OpenShift engineer
I want the shared resource CSI Driver webhook to be installed with the cluster storage operator
So that the webhook is deployed when the CSI driver is deployed

Acceptance Criteria

  • Shared Resource CSI Driver operator deploys the webhook alongside the CSI driver
  • Cluster storage operator is updated if needed to deploy the shared resource CSI driver webhook.

Docs Impact

None - no new functional capabilities will be added

QE Impact

None - we can verify in CI that we are deploying the webhook correctly.

PX Impact

None - no new functional capabilities will be added

Notes

The scope of this story is to just deploy the "hello world" webhook with the Cluster Storage Operator.
Adding the live ValidatingWebhook configuration and service will be done in a separate story.

User Story

As an OpenShift engineer,
I want to initialize a validating admission webhook for the shared resource CSI driver
So that I can eventually require readOnly: true to be set on all pods that use the Shared Resource CSI Driver

Acceptance Criteria

  • Container image created in CI which builds a "hello world" binary for the future validating webhook.
  • ART sets up downstream build process for the image.

QE Impact

None.

Docs Impact

None.

PX Impact

None.

Notes

This is a prerequisite for implementing the validating admission webhook.
We need to have ART build the container image downstream so that we can add the correct image references for the CVO.
If we reference images in the CVO manifests which do not have downstream counterparts, we break the downstream build for the payload.

CI is capable of producing multiple images for a GitHub repository. For example, github.com/openshift/oc produces 4-5 images with various capabilities.

We did similar work in BUILD-234 - some of these steps are not required.

See also:

Upstream K8s deprecated PodSecurityPolicy and replaced it with a new built-in admission controller that enforces the Pod Security Standards (See here for the motivations for deprecation).] There is an OpenShift-specific dedicated pod admission system called Security Context Constraints. Our aim is to keep the Security Context Constraints pod admission system while also allowing users to have access to the Kubernetes Pod Security Admission. 

With OpenShift 4.11, we are turned on the Pod Security Admission with global "privileged" enforcement. Additionally we set the "restricted" profile for warnings and audit. This configuration made it possible for users to opt-in their namespaces to Pod Security Admission with the per-namespace labels. We also introduced a new mechanism that automatically synchronizes the Pod Security Admission "warn" and "audit" labels.

With OpenShift 4.15, we intend to move the global configuration to enforce the "restricted" pod security profile globally. With this change, the label synchronization mechanism will also switch into a mode where it synchronizes the "enforce" Pod Security Admission label rather than the "audit" and "warn". 

Epic Goal

Get Pod Security admission to be run in "restricted" mode globally by default alongside with SCC admission.

When creating a custom SCC, it is possible to assign a priority that is higher than existing SCCs. This means that any SA with access to all SCCs might use the higher priority custom SCC, and this might mutate a workload in an unexpected/unintended way.

To protect platform workloads from such an effect (which, combined with PSa, might result in rejecting the workload once we start enforcing the "restricted" profile) we must pin the required SCC to all workloads in platform namespaces (openshift-, kube-, default).

Each workload should pin the SCC with the least-privilege, except workloads in runlevel 0 namespaces that should pin the "privileged" SCC (SCC admission is not enabled on these namespaces, but we should pin an SCC for tracking purposes).

The following table tracks progress.

= completed

# namespace 4.17 4.16 4.15
1 oc debug node pods #1763 #1816 #1818
2 openshift-apiserver-operator #573 #581
3 openshift-authentication #656 #675
4 openshift-authentication-operator #656 #675
5 openshift-catalogd #50  
6 openshift-cloud-credential-operator #681  
7 openshift-cloud-network-config-controller #2282    
8 openshift-cluster-csi-drivers #170 #459 #484
9 openshift-cluster-node-tuning-operator #968  
10 openshift-cluster-olm-operator #54  
11 openshift-cluster-samples-operator #535  
12 openshift-cluster-storage-operator #459 #196 #484 #211
13 openshift-cluster-version #1038 #1068
14 openshift-config-operator #410 #420
15 openshift-console #871 #908  
16 openshift-console-operator #871 #908  
17 openshift-controller-manager #336 #361
18 openshift-controller-manager-operator #336 #361
19 openshift-e2e-loki      
20 openshift-image-registry #1008 #1067
21 openshift-ingress #1031    
22 openshift-ingress-canary #1031    
23 openshift-ingress-operator #1031    
24 openshift-insights #915 #967
25 openshift-kni-infra      
26 openshift-kube-storage-version-migrator #107 #112
27 openshift-kube-storage-version-migrator-operator #107 #112
28 openshift-machine-api #407 #315 #282 #1220 #73 #50 #433  
29 openshift-machine-config-operator #4219 #4384 #4393
30 openshift-manila-csi-driver      
31 openshift-marketplace #561  
32 openshift-metallb-system      
33 openshift-monitoring #2335  
34 openshift-network-diagnostics #2282    
35 openshift-network-node-identity #2282    
36 openshift-nutanix-infra      
37 openshift-oauth-apiserver #656 #675
38 openshift-openstack-infra      
39 openshift-operator-controller #100  
40 openshift-operator-lifecycle-manager #703  
41 openshift-route-controller-manager #336 #361
42 openshift-rukpak      
43 openshift-service-ca #235 #243
44 openshift-service-ca-operator #235 #243
45 openshift-sriov-network-operator      
46 openshift-user-workload-monitoring #2335  
47 openshift-vsphere-infra      
48 (runlevel) kube-system      
49 (runlevel) openshift-cloud-controller-manager      
50 (runlevel) openshift-cloud-controller-manager-operator      
51 (runlevel) openshift-cluster-api      
52 (runlevel) openshift-cluster-machine-approver      
53 (runlevel) openshift-dns      
54 (runlevel) openshift-dns-operator      
55 (runlevel) openshift-etcd      
56 (runlevel) openshift-etcd-operator      
57 (runlevel) openshift-kube-apiserver      
58 (runlevel) openshift-kube-apiserver-operator      
59 (runlevel) openshift-kube-controller-manager      
60 (runlevel) openshift-kube-controller-manager-operator      
61 (runlevel) openshift-kube-proxy      
62 (runlevel) openshift-kube-scheduler      
63 (runlevel) openshift-kube-scheduler-operator      
64 (runlevel) openshift-multus      
65 (runlevel) openshift-network-operator      
66 (runlevel) openshift-ovn-kubernetes      
67 (runlevel) openshift-sdn      

Feature Overview

Telecommunications providers continue to deploy OpenShift at the Far Edge. The acceleration of this adoption and the nature of existing Telecommunication infrastructure and processes drive the need to improve OpenShift provisioning speed at the Far Edge site and the simplicity of preparation and deployment of Far Edge clusters, at scale.

Goals

  • Simplicity The folks preparing and installing OpenShift clusters (typically SNO) at the Far Edge range in technical expertise from technician to barista. The preparation and installation phases need to be reduced to a human-readable script that can be utilized by a variety of non-technical operators. There should be as few steps as possible in both the preparation and installation phases.
  • Minimize Deployment Time A telecommunications provider technician or brick-and-mortar employee who is installing an OpenShift cluster, at the Far Edge site, needs to be able to do it quickly. The technician has to wait for the node to become in-service (CaaS and CNF provisioned and running) before they can move on to installing another cluster at a different site. The brick-and-mortar employee has other job functions to fulfill and can't stare at the server for 2 hours. The install time at the far edge site should be in the order of minutes, ideally less than 20m.
  • Utilize Telco Facilities Telecommunication providers have existing Service Depots where they currently prepare SW/HW prior to shipping servers to Far Edge sites. They have asked RH to provide a simple method to pre-install OCP onto servers in these facilities. They want to do parallelized batch installation to a set of servers so that they can put these servers into a pool from which any server can be shipped to any site. They also would like to validate and update servers in these pre-installed server pools, as needed.
  • Validation before Shipment Telecommunications Providers incur a large cost if forced to manage software failures at the Far Edge due to the scale and physical disparate nature of the use case. They want to be able to validate the OCP and CNF software before taking the server to the Far Edge site as a last minute sanity check before shipping the platform to the Far Edge site.
  • IPSec Support at Cluster Boot Some far edge deployments occur on an insecure network and for that reason access to the host’s BMC is not allowed, additionally an IPSec tunnel must be established before any traffic leaves the cluster once its at the Far Edge site. It is not possible to enable IPSec on the BMC NIC and therefore even OpenShift has booted the BMC is still not accessible.

Requirements

  • Factory Depot: Install OCP with minimal steps
    • Telecommunications Providers don't want an installation experience, just pick a version and hit enter to install
    • Configuration w/ DU Profile (PTP, SR-IOV, see telco engineering for details) as well as customer-specific addons (Ignition Overrides, MachineConfig, and other operators: ODF, FEC SR-IOV, for example)
    • The installation cannot increase in-service OCP compute budget (don't install anything other that what is needed for DU)
    • Provide ability to validate previously installed OCP nodes
    • Provide ability to update previously installed OCP nodes
    • 100 parallel installations at Service Depot
  • Far Edge: Deploy OCP with minimal steps
    • Provide site specific information via usb/file mount or simple interface
    • Minimize time spent at far edge site by technician/barista/installer
    • Register with desired RHACM Hub cluster for ongoing LCM
  • Minimal ongoing maintenance of solution
    • Some, but not all telco operators, do not want to install and maintain an OCP / ACM cluster at Service Depot
  • The current IPSec solution requires a libreswan container to run on the host so that all N/S OCP traffic is encrypted. With the current IPSec solution this feature would need to support provisioning host-based containers.

 

A list of specific needs or objectives that a Feature must deliver to satisfy the Feature. Some requirements will be flagged as MVP. If an MVP gets shifted, the feature shifts.  If a non MVP requirement slips, it does not shift the feature.

requirement Notes isMvp?
     
     
     

 

Describe Use Cases (if needed)

Telecommunications Service Provider Technicians will be rolling out OCP w/ a vDU configuration to new Far Edge sites, at scale. They will be working from a service depot where they will pre-install/pre-image a set of Far Edge servers to be deployed at a later date. When ready for deployment, a technician will take one of these generic-OCP servers to a Far Edge site, enter the site specific information, wait for confirmation that the vDU is in-service/online, and then move on to deploy another server to a different Far Edge site.

 

Retail employees in brick-and-mortar stores will install SNO servers and it needs to be as simple as possible. The servers will likely be shipped to the retail store, cabled and powered by a retail employee and the site-specific information needs to be provided to the system in the simplest way possible, ideally without any action from the retail employee.

 

Out of Scope

Q: how challenging will it be to support multi-node clusters with this feature?

Background, and strategic fit

< What does the person writing code, testing, documenting need to know? >

Assumptions

< Are there assumptions being made regarding prerequisites and dependencies?>

< Are there assumptions about hardware, software or people resources?>

Customer Considerations

< Are there specific customer environments that need to be considered (such as working with existing h/w and software)?>

< Are there Upgrade considerations that customers need to account for or that the feature should address on behalf of the customer?>

<Does the Feature introduce data that could be gathered and used for Insights purposes?>

Documentation Considerations

< What educational or reference material (docs) is required to support this product feature? For users/admins? Other functions (security officers, etc)? >

< What does success look like?>

< Does this feature have doc impact?  Possible values are: New Content, Updates to existing content,  Release Note, or No Doc Impact>

< If unsure and no Technical Writer is available, please contact Content Strategy. If yes, complete the following.>

  • <What concepts do customers need to understand to be successful in [action]?>
  • <How do we expect customers will use the feature? For what purpose(s)?>
  • <What reference material might a customer want/need to complete [action]?>
  • <Is there source material that can be used as reference for the Technical Writer in writing the content? If yes, please link if available. >
  • <What is the doc impact (New Content, Updates to existing content, or Release Note)?>

Interoperability Considerations

< Which other products and versions in our portfolio does this feature impact?>

< What interoperability test scenarios should be factored by the layered product(s)?>

Questions

Question Outcome
   

 

 

Epic Goal

  • Install SNO within 10 minutes

Why is this important?

  • SNO installation takes around 40+ minutes.
  • This makes SNO less appealing when compared to k3s/microshift.
  • We should analyze the  SNO installation, figure our why it takes so long and come up with ways to optimize it

Scenarios

  1. ...

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • ...

Dependencies (internal and external)

  1. ...

Previous Work (Optional):

Open questions::

  1. https://docs.google.com/document/d/1ULmKBzfT7MibbTS6Sy3cNtjqDX1o7Q0Rek3tAe1LSGA/edit?usp=sharing

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

openshift- service-ca service-ca pod takes a few minutes to start when installing SNO

kubectl get events -n openshift-service-ca --sort-by='.metadata.creationTimestamp' -o custom-columns=FirstSeen:.firstTimestamp,LastSeen:.lastTimestamp,Count:.count,From:.source.component,Type:.type,Reason:.reason,Message:.message                      
FirstSeen              LastSeen               Count   From                                                                                              Type      Reason                 Message
2023-01-22T12:25:58Z   2023-01-22T12:25:58Z   1       deployment-controller                                                                             Normal    ScalingReplicaSet      Scaled up replica set service-ca-6dc5c758d to 1
2023-01-22T12:26:12Z   2023-01-22T12:27:53Z   9       replicaset-controller                                                                             Warning   FailedCreate           Error creating: pods "service-ca-6dc5c758d-" is forbidden: error fetching namespace "openshift-service-ca": unable to find annotation openshift.io/sa.scc.uid-range
2023-01-22T12:27:58Z   2023-01-22T12:27:58Z   1       replicaset-controller                                                                             Normal    SuccessfulCreate       Created pod: service-ca-6dc5c758d-k7bsd
2023-01-22T12:27:58Z   2023-01-22T12:27:58Z   1       default-scheduler                                                                                 Normal    Scheduled              Successfully assigned openshift-service-ca/service-ca-6dc5c758d-k7bsd to master1
 

Seems that creating the serivce-ca namespace early allows it to get
openshift.io/sa.scc.uid-range annotation and start running earlier, the
service-ca pod is required for other pods (CVO and all the control plane pods) to start since it's creating the serving-cert 

Complete Epics

This section includes Jira cards that are linked to an Epic, but the Epic itself is not linked to any Feature. These epics were completed when this image was assembled

Summary (PM+lead)

https://issues.redhat.com/browse/AUTH-2 revealed that, in prinicipal, Pod Security Admission is possible to integrate into OpenShift while retaining SCC functionality.

 

This epic is about the concrete steps to enable Pod Security Admission by default in OpenShift

Motivation (PM+lead)

Goals (lead)

  • Enable Pod Security Admission in "restricted" policy level by default
  • Migrate existing core workloads to comply to the "restricted" pod security policy level

Non-Goals (lead)

  • Other OpenShift workloads must be migrated by the individual responsible teams.

Deliverables

Proposal (lead)

Enhancement - https://github.com/openshift/enhancements/pull/1010

User Stories (PM)

Dependencies (internal and external, lead)

Previous Work (lead)

Open questions (lead)

  1. ...

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

dns-operator must comply to restricted pod security level. The current audit warning is:

{   "objectRef": "openshift-dns-operator/deployments/dns-operator",   "pod-security.kubernetes.io/audit-violations": "would violate PodSecurity \"restricted:latest\": allowPrivilegeEscalation != false (containers \"dns-operator\", \"kube-rbac-proxy\" must set securityContext.allowPrivilegeEscalation=false), unre stricted capabilities (containers \"dns-operator\", \"kube-rbac-proxy\" must set securityContext.capabilities.drop=[\"ALL\"]), runAsNonRoot != true (pod or containers \"dns-operator\", \"kube-rbac-proxy\" must set securityContext.runAsNonRoot=tr ue), seccompProfile (pod or containers \"dns-operator\", \"kube-rbac-proxy\" must set securityContext.seccompProfile.type to \"RuntimeDefault\" or \"Localhost\")" }

ingress-operator must comply to pod security. The current audit warning is:

 

{   "objectRef": "openshift-ingress-operator/deployments/ingress-operator",   "pod-security.kubernetes.io/audit-violations": "would violate PodSecurity \"restricted:latest\": allowPrivilegeEscalation != false (containers \"ingress-operator\", \"kube-rbac-proxy\" must set securityContext.allowPrivilegeEscalation=false), unrestricted capabilities (containers \"ingress-operator\", \"kube-rbac-proxy\" must set securityContext.capabilities.drop=[\"ALL\"]), runAsNonRoot != true (pod or containers \"ingress-operator\", \"kube-rbac-proxy\" must set securityContext.run AsNonRoot=true), seccompProfile (pod or containers \"ingress-operator\", \"kube-rbac-proxy\" must set securityContext.seccompProfile.type to \"RuntimeDefault\" or \"Localhost\")" }

Epic Goal

HyperShift provisions OpenShift clusters with externally managed control-planes. It follows a slightly different process for provisioning clusters. For example, HyperShift uses cluster API as a backend and moves all the machine management bits to the management cluster.  

Why is this important?

showing machine management/cluster auto-scaling tabs in the console is likely to confuse users and cause unnecessary side effects. 

Definition of Done

  • MachineConfig and MachineConfigPool should not be present, they should be either removed or hidden when the cluster is spawned using HyperShift. 
  • Cluster Settings show say the control plane is externally managed and be read-only.
  • Cluster Settings -> Configuration resources should be read-only, maybe hide the tab
  • Some resources should go in an allowlist. Most will be hidden
  • Review getting started steps

See Design Doc: https://docs.google.com/document/d/1k76JtRRHBdCCEjHPqKcYvbNVsuaGmRhWDLESWIm0mbo/edit#

 

Setup / Testing

It's based on the SERVER_FLAG controlPlaneTopology being set to External is really the driving factor here; this can be done in one of two ways:

  • Locally via a Bridge Variable, export BRIDGE_CONTROL_PLANE_TOPOLOGY_MODE="External"
  • Locally / OnCluster via modifying the window.SERVER_FLAGS.controlPlaneTopology to External in the dev tools

To test work related to cluster upgrade process, use a 4.10.3 cluster set on the candidate-4.10 upgrade channel using 4.11 frontend code.

If the Infrastructure.Status.ControlPlaneTopology is set to 'External', the console-operator will pass this information via the console-config.yaml co the console. Console pod will get re-deployed and will store the topology mode information as a SERVER_FLAG. Based on that value we need surface a message that the control plane is externally managed and add following changes:

  • Remove update button
  • Make channel read only
  • Link out to read only CV details page
  • Remove the ability to edit upstream configuration
  • Remove the cluster autoscaler field
  • Add an alert to the page so that users know the control plane is externally managed

In general, anything that changes a cluster version should be read only.

Check section 02 for more info: https://docs.google.com/document/d/1k76JtRRHBdCCEjHPqKcYvbNVsuaGmRhWDLESWIm0mbo/edit#

 

If the Infrastructure.Status.ControlPlaneTopology is set to 'External', the console-operator will pass this information via the console-config.yaml to the console. Console pod will get re-deployed and will store the topology mode information as a SERVER_FLAG. Based on that value we need to suspend kubeadmin notifier, from the global notifications, since it contain link for updating the cluster OAuth configuration (see attachment).

 

 

If the Infrastructure.Status.ControlPlaneTopology is set to 'External', the console-operator will pass this information via the console-config.yaml to the console. Console pod will get re-deployed and will store the topology mode information as a SERVER_FLAG. Based on that value we need to suspend these notifications:

  • cluster upgrade notifications
  • new channel available notifications

For these we will need to check `ControlPlaneTopology`, if it's set to 'External' and also check if the user can edit cluster version(either by creating a hook or an RBAC call, eg. `canEditClusterVersion`)

 

Check section 05 for more info: https://docs.google.com/document/d/1k76JtRRHBdCCEjHPqKcYvbNVsuaGmRhWDLESWIm0mbo/edit#

Based on Cesar's comment we should be removing the `Control Plane` section, if the infrastructure.status.controlplanetopology being "External".

If the Infrastructure.Status.ControlPlaneTopology is set to 'External', the console-operator will pass this information via the console-config.yaml co the console. Console pod will get re-deployed and will store the topology mode information as a SERVER_FLAG. Based on that value we need to remove the ability to “Add identity providers” under “Set up your Cluster”. In addition to the getting started card, we should remove the ability to update a cluster on the details card when applicable (anything that changes a cluster version should be read only).

Summary of changes to the overview page:

  • Remove the ability to “Add identify providers” under “Set up your Cluster”
  • Remove cluster update CTA from the details card
  • Remove update alerts from the status card

Check section 03 for more info: https://docs.google.com/document/d/1k76JtRRHBdCCEjHPqKcYvbNVsuaGmRhWDLESWIm0mbo/edit#

Epic Goal

Why is this important?

  • So the UX satisfies the current trands, where dark mode is becoming a standard for modern services.  

Acceptance Criteria

  • OCP admin console must be rendered in a preferred mode based on `prefers-color-scheme` media query
  • OCP admin console must be rendered in a preferred mode selected in the User Setting page
  • Create an followup epic/story for and listing and tracking changes needed in OCP console's dynamic plugins

Dependencies (internal and external)

  1. PatternFly - Dark mode PF variables

Previous Work (Optional):

  1. Mike Coker has worked on a POC from the PF point of view on both the admin and dev console, and the screenshot results are listed below along with the repo branch. Also listed is a document covering some of the common issues found when putting together the admin console POC. https://github.com/mcoker/console/tree/dark-theme
    Background POC work completed for reference:

PatternFly Dark Theme Handbookhttps://docs.google.com/document/d/1mRYEfUoOjTsSt7hiqjbeplqhfo3_rVDO0QqMj2p67pw/edit

Admin Console -> Workloads & Pods

Dev Console -> Gotcha pages: Observe Dashboard and Metrics, Add, Pipelines: builder, list, log, and run

Open questions::

  1. Who should be responsible for updating DynamicPlugins to be able to render in dark mode?

As a developer, I want to be able to fix remaining issues from the spreadsheet of issues generated after the initial pass and spike of adding dark theme to the console.. As such, I need to make sure to either complete all remaining issues for the spreadsheet, or, create a bug or future story for any remaining issues in these two documents.

 

Acceptance criteria:

As a developer, I want to be able to scope the changes needed to enable dark mode for the admin console. As such, I need to investigate how much of the console will display dark mode using PF variables and also define a list of gotcha pages/components which will need special casing above and beyond PF variable settings.

 

Acceptance criteria:

An epic we can duplicate for each release to ensure we have a place to catch things we ought to be doing regularly but can tend to fall by the wayside.

The Cluster Dashboard Details Card Protractor integration test was failing at high rate, and despite multiple attempts to fix, was never fully resolved, so it was disabled as a way to fix https://bugzilla.redhat.com/show_bug.cgi?id=2068594. Migrating this entire file to Cypress should give us better debugging capability, which is what was done to fix a similarly problematic project dashboard Protractor test.

This epic contains all the Dynamic Plugins related stories for OCP release-4.11 

Epic Goal

  • Track all the stories under a single epic

Acceptance Criteria

  •  

We have a Timestamp component for consistent display of dates and times that we should expose through the SDK. We might also consider a hook that formats dates and times for places were you don't want or cant use the component, eg. times on a chart. 

This will become important when we add a user preference for dates so that plugins show consistent dates and times as console. If I set my user preference to UTC dates, console should show UTC dates everywhere.

 

AC:

  • Expose the Timestamp component inside the SDK. 
  • Replace the connect with useSelector hook
  • Keep the original component and proxy it to the new one in the SDK

 

 

 

cc Jakub Hadvig Sho Weimer 

In the 4.11 release, a console.openshift.io/default-i18next-namespace annotation is being introduced. The annotation indicates whether the ConsolePlugin contains localization resources. If the annotation is set to "true", the localization resources from the i18n namespace named after the dynamic plugin (e.g. plugin__kubevirt), are loaded. If the annotation is set to any other value or is missing on the ConsolePlugin resource, localization resources are not loaded. 

 

In case these resources are not present in the dynamic plugin, the initial console load will be slowed down. For more info check BZ#2015654

 

AC:

  • console-operator should be checking for the new console.openshift.io/use-i18n annotation, update the console-config.yaml accordingly and redeploy the console server
  • console server should pick up the changes in the console-config.yaml and only load the i18n namespace that are available

 

Follow up of https://issues.redhat.com/browse/CONSOLE-3159

 

 

We need to provide a base for running integration tests using the dynamic plugins. The tests should initially

  • Create a deployment and service to run the dynamic demo plugin
  • Update the console operator config to enable the plugin
  • Wait for the plugin to be available
  • Test at least one extension point used by the plugin (such as adding items to the nav)
  • Disable the plugin when done

Once the basic framework is in place, we can update the demo plugin and add new integration tests when we add new extension points.

https://github.com/openshift/console/tree/master/frontend/dynamic-demo-plugin

 

https://github.com/openshift/enhancements/blob/master/enhancements/console/dynamic-plugins.md

 

https://github.com/openshift/console/tree/master/frontend/packages/console-plugin-sdk

Currently, enabled plugins can fail to load for a variety of reasons. For instance, plugins don't load if the plugin name in the manifest doesn't match the ConsolePlugin name or the plugin has an invalid codeRef. There is no indication in the UI that something has gone wrong. We should explore ways to report this problem in the UI to cluster admins. Depending on the nature of the issue, an admin might be able to resolve the issue or at least report a bug against the plugin.

The message about failing could appear in the notification drawer and/or console plugins tab on the operator config. We could also explore creating an alert if a plugin is failing.

 

AC:

  • Add notification into the Notification Drawer in case a Dynamic Plugin will error out during load.
  • Render these errors in the status card, notification section, as well.
  • For each failed plugin we should create a separate notification.

Currently, you need to navigate to

Cluster Settings ->
Global configuration ->
Console (operator) config ->
Console plugins

to see and managed plugins. This takes a lot of clicks and is not discoverable. We should look at surfacing plugin details where they're easier to find – perhaps on the Cluster Settings page – or at least provide a more convenient link somewhere in the UI.

AC: Add the Dynamic Plugins section to the Status Card in the overview that will contain:

  • count of active and non-active plugins
  • link to the ConsolePlugins instances page
  • status of the loaded plugins and breakout error

cc Ali Mobrem Robb Hamilton

Goal

  • Add the ability for users to select supported but not recommended updates.
  • Refine workflow when both "upgradeable=false" and "supported-but-not-recommended" updates occur

Background
RFE: for 4.10, Cincinnati and the cluster-version operator are adding conditional updates (a.k.a. targeted edge blocking): https://issues.redhat.com/browse/OTA-267

High-level plans in https://github.com/openshift/enhancements/blob/master/enhancements/update/targeted-update-edge-blocking.md#update-client-support-for-the-enhanced-schema

Example of what the oc adm upgrade UX will be in https://github.com/openshift/enhancements/blob/master/enhancements/update/targeted-update-edge-blocking.md#cluster-administrator.

The oc implementation landed via https://github.com/openshift/oc/pull/961.

Design

  • Use case 01: "supported but not recommended" occurs to the latest version:
    • Add an info icon next to the version on update path with a pop-over to explain about why updating to this version is supported, but not recommended and a link to known risks
    • Identify the difference in "recommended" versions, "supported but not recommended" versions, and "blocked" versions (upgradeable=false) in the + more modal.
    • The latest version is pre-selected in the dropdown in the update modal with an inline alert to inform users about supported-but-not-recommended version with link to known risks. Users can choose to update to another recommended versions, update to a supported-but-not-recommended one, or wait.
    • The "recommended" and "supported but not recommended" updates are separated in the dropdown.
    • If a user selects a "recommended" update, the inline alert disappears.
  • Use case 02: When both "upgradeable=false" and "supported but not recommended" occur:
    • Add an alert banner to explain why users shouldn’t update to the latest version and link to how to resolve on the cluster settings details page. Users have the options to resolve the issue, update to a patch version, or wait.
    • If users open the update modal without resolving the "upgradeable=false" issue, the next recommended version is pre-selected. An expandable link "View blocked versions (#)" is included under the dropdown to show "upgradeable=false" versions with resolve link.
    • If users resolve the "upgradeable=false" issue, the cluster settings page will change to use case 01
    • Question: Priority on changing the upgradeable=false alert banner in update modal and blocked versions in dropdown

See design doc: https://docs.google.com/document/d/1Nja4whdsI5dKmQNS_rXyN8IGtRXDJ8gXuU_eSxBLMIY/edit#

See marvel: https://marvelapp.com/prototype/h3ehaa4/screen/86077932

Update the cluster settings page to inform the user when the latest available update is supported but not recommended. Add an informational popover to the latest version in  update path visualization.

The "Update Version" modal on the cluster settings page should be updated to give users information about recommended, not recommended, and blocked update versions.

  • When the modal is opened, the latest recommended update version should be pre-selected in the version dropdown.
  • Blocked versions should no longer be displayed in the version dropdown, and should instead be displayed in a collapsible field below the dropdown.
  • When blocked versions are present, a link should be provided to the cluster operator tab. The version dropdown itself should have two labeled sections: "Recommended" and "Supported but not recommended".
  • When the user selects a "Supported but not recommended" item from the version dropdown, an inline info alert should appear below the version selection field and should provide a link to known risks associated with the selected version. This is an external link provided through the ClusterVersion API.

Epic Goal

  • Add telemetry so that we know how image stream features are used.

Why is this important?

  • We have a long standing epic to create image streams v2. We need to better understand how image streams are used today.

Scenarios

  1. ...

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • ...

Dependencies (internal and external)

  1. ...

Previous Work (Optional):

Open questions::

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Epic Goal

  • Make the image registry distributed across availability zones.

Why is this important?

  • The registry should be highly available and zone failsafe.

Scenarios

  1. As an administrator I want to rely on a default configuration that spreads image registry pods across topology zones so that I don't suffer from a long recovery time (>6 mins) in case of a complete zone failure if all pods are impacted.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.

Dependencies (internal and external)

  1. Pod's topologySpreadConstraints

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: https://github.com/openshift/cluster-image-registry-operator/pull/730
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Story: As an administrator I want to rely on a default configuration that spreads image registry pods across topology zones so that I don't suffer from a long recovery time (>6 mins) in case of a complete zone failure if all pods are impacted.

Background: The image registry currently uses affinity/anti-affinity rules to spread registry pods across different hosts. However this might cause situations in which all pods end up on hosts of a single zone, leading to a long recovery time of the registry if that zone is lost entirely. However due to problems in the past with the preferred setting of anti-affinity rule adherence the configuration was forced instead with required and the rules became constraints. With zones as constraints the internal registry would not have deployed anymore in environments with a single zone, e.g. internal CI environment. Pod topology constraints is a new API that is supported in OCP which can also relax constraints in case they cannot be satisfied. Details here: https://docs.openshift.com/container-platform/4.7/nodes/scheduling/nodes-scheduler-pod-topology-spread-constraints.html

Acceptance criteria:

  • by default the internal registry is deployed with at least two replica
  • by default the topology constraints should be on a zone-basis, so that by defaults one registry pod is scheduled in each zone
  • when constraints can't be satisfied the registry should deploy anyway
  • we should not do this in SNO environments
  • the registry should still work on SNO environments

Open Questions:

  • what happens in environments where the storage is zone dependent?
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

As an OpenShift administrator
I want to provide the registry operator with a custom certificate authority for S3 storage
so that I can use a third-party S3 storage provider.

Acceptance criteria

  1. Users can specify a configmap name (from openshift-config) in config.imageregistry/cluster's spec.storage.s3.
  2. The operator uses CA from this configmap to check S3 bucket.
  3. The image registry pod uses CA from this configmap to access the S3 bucket.
  4. When a custom CA is defined, the operator/image-registry should still trust certificate authorities that are used by Amazon S3 and other well-known CAs.
  5. An end-to-end test that runs minio and checks the image registry becomes healthy with it.

Goal

Remove Jenkins from the OCP Payload.

Problem

  • Jenkins images are "non-trival in size, impact experience around OCP payloads
  • Security advisories cannot be handled once, but against all actively supported OCP releases, adding to response time for handling said advisories
  • Some customers may now want to upgrade Jenkins as OCP upgrades (making this configurable is more ideal)

Why is this important

  • This is an engineering motivated item to reduce costs so we have more cycles for strategic work
  • Aside from the team itself, top level OCP architects want this to reduce the image size, improve general OCP upgrade experience
  • Sends a mix message with respect to what is startegic CI/CI when Jenkins is baked into OCP, but Tekton/Pipelines is an add-on, day 2 install sort of thing

Dependencies (internal and external)

See epic linking - need alternative non payload image available to provide relatively seamless migration

 

Also, the EP for this is approved and merged at https://github.com/openshift/enhancements/blob/master/enhancements/builds/remove-jenkins-payload.md

Estimate (xs, s, m, l, xl, xxl):

Questions:

       PARTIAL ANSWER ^^:  confirmed with Ben Parees in https://coreos.slack.com/archives/C014MHHKUSF/p1646683621293839 that EP merging is currently sufficient OCP "technical leadership" approval.

 

Previous work

 

Customers

assuming none

User Stories

 

As maintainers of the OpenShift jenkins component, we need run Jenkins CI for PR testing against openshift/jenkins, openshift/jenkins-sync-plugin, openshift/jenkins-client-plugin, openshift/jenkins-openshift-login-plugin, using images built in the CI pipeline but not injected into CI test clusters via sample operator overriding the jenkins sample imagestream with the jenkins payload image.

 

As maintainers of the OpenShift Jenkins component, we need Jenkins periodics for the client and sync plugins to run against the latest non payload, CPaas image, promoted to CI's image locations on quay.io, for the current release in development.

 

As maintainers of the OpenShift Jenkins component, we need Jenkins related tests outside of very basic Jenkins Pipieline Strategy Build Config verification, removed from openshift-tests in OpenShift Origin, using a non-payload, CPaas image pertinent to the branch in question.

Acceptance criteria

  • all PR CI Tests do not utilize samples operator manipulation of the jenkins imagestream with the in payload image, but rather images including the PRs changes
  • all periodic CI Tests do not utilize samples operator manipulation of the jenkins imagestream with the in payload image, but rather CI promoted images for the current release pushed to quay.io

High Level, we ideally want to vet the new CPaas image via CI and periodics BEFORE we start changing the samples operator so that it does not manipulate the jenkins imagestream (our tests will override the samples operator override)

QE Impact

NONE ... QE should wait until JNKS-254

Docs Impact

NONE

PX Impact

 

NONE

Launch Checklist

Dependencies identified
Blockers noted and expected delivery timelines set
Design is implementable
Acceptance criteria agreed upon
Story estimated

Notes

  • Our CSI shared resource experience will help us here
  • but the old IMAGE_FORMAT stuff is deprecated, and does not work well with step registry stuff
  • instead, we need to use https://docs.ci.openshift.org/docs/architecture/ci-operator/#dependency-overrides
  • Makefile level logic will use `oc tag` to update the jenkins imagestream created as part of samples to override the use of the in payload image with the image build by the PR, or for periodics, with what has been promoted to quay.io
  • Ultimately, CI step registry for capturing the `oc tag` update the imagestream logic is the probably end goal
  • JNKS-268 might change how we do periodics, but the current thought is to get existing periodics working with the CPaas image first

Possible staging

1) before CPaas is available, we can validate images generated by PRs to openshift/jenkins, openshift/jenkins-sync-plugin, openshift/jenkins-client-plugin by taking the image built by the image (where the info needed to get the right image from the CI registry is in the IMAGE_FORMAT env var) and then doing an `oc tag --source=docker <PR image ref> openshift/jenkins:2` to replace the use of the payload image in the jenkins imagestream in the openshift namespace with the PRs image

2) insert 1) in https://github.com/openshift/release/blob/master/ci-operator/step-registry/jenkins/sync-plugin/e2e/jenkins-sync-plugin-e2e-commands.sh and https://github.com/openshift/release/blob/master/ci-operator/step-registry/jenkins/client-plugin/tests/jenkins-client-plugin-tests-commands.sh where you test for IMAGE_FORMAT being set

3) or instead of 2) you update the Makefiles for the plugins to call a script that does the same sort of thing, see what is in IMAGE_FORMAT, and if it has something, do the `oc tag`

 

https://github.com/openshift/release/pull/26979 is a prototype of how to stick the image built from a PR and conceivably the periodics to get the image built from it and tag it into the jenkins imagestream in the openshift namespace in the test cluster

 

Epic Goal

  • Provide a dedicated dashboard for NVIDIA GPU usage visualization in the OpenShift Console.

Why is this important?

  • Customers that use GPUs in their clusters usually have the GPU workloads as the main purpose of their cluster. As such, it makes much more sense to have the details about the usage they are doing of GPGPU resources AND CPU/RAM rather than just CPU/RAM

Scenarios

  1. As an admin of a cluster dedicated to data science, I want to quickly find out how much of my very costly resources are currently in use and if things are getting queued due to lack of resources

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.

Dependencies (internal and external)

  1. The NVIDIA GPU Operator must export to prometheus the relevant data

Open questions::

  1. Will NVIDIA agree to these extra data exports in their GPU Operator?

I asked Zvonko Kaiser and he seemed open to it. I need to confirm with Shiva Merla

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Epic Goal

  • Remove this UI from our stack that we cannot support.

Why is this important?

  • Reduce support burden.
  • Remove Bugzilla burden of addressing continuous CVEs found in this project.

Acceptance Criteria

  • All Prometheus upstream UI links are removed
  • Related documentation is updated
  • Ports/routes etc configured to expose access to this UI are removed such that no configuration we provide enables access to this UI or its codepaths.
  • There is no reason any CVEs found in this UI would ever require intervention by the Monitoring Team.

Dependencies (internal and external)

  1. Make the Prometheus Targets information available in Console UI (https://issues.redhat.com/browse/MON-1079)

Previous Work (Optional):

Open questions::

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

After installing or upgrading to the latest OCP version, the existing OpenShift route to the prometheus-k8s service is updated to be a path-based route to '/api/v1'.

DoD:

  • It is not possible to access the Prometheus UI via the OpenShift route
  • Using a bearer token with sufficient permissions, it is possible to access the /api/v1/* endpoints via the OpenShift route.
The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Following up on https://issues.redhat.com/browse/MON-1320, we added three new CLI flags to Prometheus to apply different limits on the samples' labels. These new flags are available starting from Prometheus v2.27.0, which will most likely be shipped in OpenShift 4.9.

The limits that we want to look into for OCP are the following ones:

# Per-scrape limit on number of labels that will be accepted for a sample. If
# more than this number of labels are present post metric-relabeling, the
# entire scrape will be treated as failed. 0 means no limit.
[ label_limit: <int> | default = 0 ]

# Per-scrape limit on length of labels name that will be accepted for a sample.
# If a label name is longer than this number post metric-relabeling, the entire
# scrape will be treated as failed. 0 means no limit.
[ label_name_length_limit: <int> | default = 0 ]

# Per-scrape limit on length of labels value that will be accepted for a sample.
# If a label value is longer than this number post metric-relabeling, the
# entire scrape will be treated as failed. 0 means no limit.
[ label_value_length_limit: <int> | default = 0 ]

We could benefit from them by setting relatively high values that could only induce unbound cardinality and thus reject the targets completely if they happened to breach our constrainst.

DoD:

  • Being able to configure label scrape limits for UWM

Epic Goal

When users configure CMO to interact with systems outside of an OpenShift cluster, we want to provide an easy way to add the cluster ID to the data send.

Why is this important?

Technically this can be achieved today, by adding an identifying label to the remote_write configuration for a given cluster. The operator adding the remote_write integration needs to take care that the label is unique over the managed fleet of clusters. This however adds management complexity. Any given cluster already has a pseudo-unique datum, that can be used for this purpose.

  • Starting in 4.9 we support the Prometheus remote_write feature to send metric data to a storage integration outside of the cluster similar to our own Telemetry service.
  • In Telemetry we already use the cluster ID to distinguish the various clusters.
  • For users of remote_write this could add an easy way to add such distinguishing information.

Scenarios

  1. An organisation with multiple OpenShift clusters want to store their metric data centralized in a dedicated system and use remote_write in all their clusters to send this data. When querying their centralized storage, metadata (here a label) is needed to separate the data of the various clusters.
  2. Service providers who manage multiple clusters for multiple customers via a centralized storage system need distinguishing metadata too. See https://issues.redhat.com/browse/OSD-6573 for example

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • Document how to use this feature

Dependencies (internal and external)

  1. none

Previous Work (Optional):

  1. none

Open questions::

  1.  

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Implementation proposal:

 

Expose a flag in the CMO configuration, that is false by default (keeps backward compatibility) and when set to true will add the _id label to a remote_write configuration. More specifically it will be added to the top of a remote_write relabel_config list via the replace action. This will add the label as expect, but additionally a user could alter this label in a later relabel config to suit any specific requirements (say rename the label or add additional information to the value).
The location of this flag is the remote_write Spec, so this can be set for individual remote_write configurations.

We currently use a sample app to e2e test remote write in CMO.
In order to test the addition of the cluster_id relabel config, we need to confirm that the metrics send actually have the expected label.
For this test we should use Prometheus as the remote_write target. This allows us to query the metrics send via remote write and confirm they have the expected label.

Add an optional boolean flag to CMOs definition of RemoteWriteSpec that if true adds an entry in the specs WriteRelabelConfigs list.

I went with adding the relabel config to all user-supplied remote_write configurations. This path has no risk for backwards compatibility (unless users use the {}tmp_openshift_cluster_id{} label, seems unlikely) and reduces overall complexity, as well as documentation complexity.

The entry should look like what is already added to the telemetry remote write config and it should be added as the first entry in the list, before any user supplied relabel configs.

Epic Goal

  • Offer the option to double the scrape intervals for CMO controlled ServiceMonitors in single node deployments
  • Alternatively automatically double the same scrape intervals if CMO detects an SNO setup

The potential target ServiceMonitors are:

  • kubelet
  • kube-state-metrics
  • node-exporter
  • etcd
  • openshift-state-metrics

Why is this important?

  • Reduce CPU usage in SNO setups
  • Specifically doubling the scrape interval is important because:
  1. we are confident that this will have the least chance to interfere with existing rules. We typically have rate queries over the last 2 minutes (no shorter time window). With 30 second scrape intervals (the current default) this gives us 4 samples in any 2 minute window. rate needs at least 2 samples to work, we want another 2 for failure tolerance. Doubling the scrape interval will still give us 2 samples in most 2 minute windows. If a scrape fails, a few rule evaluations might fail intermittently.
  2. We expect a measureable reduction of CPU resources (see previous work)

Scenarios

  1. RAN deployments (Telco Edge) are SNO deployments. In these setups a full CMO deployment is often not needed and the default setup consumes too many resources. OpenShift as a whole has only very limited CPU cycles available and too many cycles are spend on Monitoring

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement details and documents.
  • ...

Previous Work (Optional):

  1. https://issues.redhat.com/browse/MON-1569

Open questions:

  1. Whether doubling some scrape intervals reduces CPU usage to fit into the assigned budget

Non goals

  • Allow arbitrarily long scrape intervals. This will interfere with alert and recoring rules
  • Implement a global override to scrape intervals.

Epic Goal

  • The goal of this epic is to update our owned repos with security scans using SNYK and to update the insecure dependencies.

Why is this important?

  • This is important because it can help use improve the overall security of shipped containers and pre-detect situations that would otherwise appear as emergency security fixes

Additional Context

Acceptance Criteria

Template:

Networking Definition of Planned

Epic Template descriptions and documentation

Epic Goal

Bump OpenShift Router from Haproxy from 2.6 and 2.8. 

Why is this important?

As a cluster administrator, I want OpenShift to include a recent HAProxy version, so that I have the latest available performance and security fixes.  

Planning Done Checklist

The following items must be completed on the Epic prior to moving the Epic from Planning to the ToDo status

  • Priority+ is set by engineering
  • Epic must be Linked to a +Parent Feature
  • Target version+ must be set
  • Assignee+ must be set
  • (Enhancement Proposal is Implementable
  • (No outstanding questions about major work breakdown
  • (Are all Stakeholders known? Have they all been notified about this item?
  • Does this epic affect SD? {}Have they been notified{+}? (View plan definition for current suggested assignee)
    1. Please use the “Discussion Needed: Service Delivery Architecture Overview” checkbox to facilitate the conversation with SD Architects. The SD architecture team monitors this checkbox which should then spur the conversation between SD and epic stakeholders. Once the conversation has occurred, uncheck the “Discussion Needed: Service Delivery Architecture Overview” checkbox and record the outcome of the discussion in the epic description here.
    2. The guidance here is that unless it is very clear that your epic doesn’t have any managed services impact, default to use the Discussion Needed checkbox to facilitate that conversation.

Additional information on each of the above items can be found here: Networking Definition of Planned

Acceptance Criteria

  • CI - MUST be running successfully with tests automated
  • Release Technical Enablement - Provide necessary release enablement
    details and documents.
  • OpenShift Router is running a version of Haproxy 2.8
  • Perf & Scale analysis is complete
  • Review of critical/major bug fixes between Haproxy 2.6 and 2.8

...

Dependencies (internal and external)

1. Perf & Scale

...

Previous Work (Optional):

1. …

Open questions::

1. …

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

Push the bump and build the new HaProxy RPM after it has been Perf & Scale tested and reviewed. 

OCP/Telco Definition of Done
Epic Template descriptions and documentation.

<--- Cut-n-Paste the entire contents of this description into your new Epic --->

Epic Goal

  • Run OpenShift builds that do not execute as the "root" user on the host node.

Why is this important?

  • OpenShift builds require an elevated set of capabilities to build a container image
  • Builds currently run as root to maintain adequate performance
  • Container workloads should run as non-root from the host's perspective. Containers running as root are a known security risk.
  • Builds currently run as root and require a privileged container. See BUILD-225 for removing the privileged container requirement.

Scenarios

  1. Run BuildConfigs in a multi-tenant environment
  2. Run BuildConfigs in a heightened security environment/deployment

Acceptance Criteria

  • Developers can opt into running builds in a cri-o user namespace by providing an environment variable with a specific value.
  • When the correct environment variable is provided, builds run in a cri-o user namespace, and the build pod does not require the "privileged: true" security context.
  • User namespace builds can pass basic test scenarios for the Docker and Source strategy build.
  • Steps to run unprivileged builds are documented.

Dependencies (internal and external)

  1. Buildah supports running inside a non-privileged container
  2. CRI-O allows workloads to opt into running containers in user namespaces.

Previous Work (Optional):

  1. BUILD-225 - remove privileged requirement for builds.

Open questions::

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • Release Enablement <link to Feature Enablement Presentation>
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • DEV - Upstream documentation merged: <link to meaningful PR or GitHub Issue>
  • DEV - Downstream build attached to advisory: <link to errata>
  • QE - Test plans in Polarion: <link or reference to Polarion>
  • QE - Automated tests merged: <link or reference to automated tests>
  • DOC - Downstream documentation merged: <link to meaningful PR>

User Story

As a developer building container images on OpenShift
I want to specify that my build should run without elevated privileges
So that builds do not run as root from the host's perspective with elevated privileges

Acceptance Criteria

  • Developers can provide an environment variable to indicate the build should not use privileged containers
  • When the correct env var + value is specified, builds run in a user namespace (non-root on the host)

QE Impact

No QE required for Dev Preview. OpenShift regression testing will verify that existing behavior is not impacted.

Docs Impact

We will need to document how to enable this feature, with sufficient warnings regarding Dev Preview.

PX Impact

This likely warrants an OpenShift blog post, potentially?

Notes

The details of this Jira Card are restricted (Red Hat Employee and Contractors only)

Description

As a user, I want to understand which service bindings connected a service to a component successfully or not. Currently it's really difficult to understand and needs inspection into each ServiceBinding resource (yaml).

Acceptance Criteria

  1. Show a status badge on the SB details page
  2. Show a Status field in the right column of the SB details page
  3. Show the Status field in the right column of the Topology side panel when a SB is selected
  4. Show an indicator in the Topology view which will help to differentiate when the service binding is in error state
  5. Define the available statuses & associated icons 🥴
    1. Connected
    2. Error
  6. Error states defined by the SB conditions … if any of these 3 are not True, the status will be displayed as Error

Additional Details:

See also https://docs.google.com/document/d/1OzE74z2RGO5LPjtDoJeUgYBQXBSVmD5tCC7xfJotE00/edit

Description

As a user, I want the topology view to be less cluttered as I doom out showing only information that I can discern and still be able to get a feel for the status of my project.

Acceptance Criteria

  1. When zoomed to 50% scale, all labels & decorators will be hidden. Label are shown when hovering over the node
  2. When zoomed to 30% scale, all labels, decorators, pod rings & icons will be hidden. Node shape remains the same, and background is either white, yellow or red. Background color is determined based on aggregate status of pods, alerts, builds and pipelines. Tooltip is available showing node name as well as the "things" which are attributing to the warning/error status.

Additional Details:

Problem:

This epic is mainly focused on the 4.10 Release QE activities

Goal:

1. Identify the scenarios for automation
2. Segregate the test Scenarios into smoke, Regression and other user stories
a. Update the https://docs.jboss.org/display/ODC/Automation+Status+Report
3. Align with layered operator teams for updating scripts
3. Work closely with dev team for epic automation
4. Create the automation scripts using cypress
5. Implement CI for nightly builds
6. Execute scripts on sprint basis

Why is it important?

To the track the QE progress at one place in 4.10 Release Confluence page

Use cases:

  1. <case>

Acceptance criteria:

  1. <criteria>

Dependencies (External/Internal):

Design Artifacts:

Exploration:

Note:

Acceptance criteria:

  1. Execute the automation scripts on ODC nightly builds in OpenShift CI (prow) periodically
  2. provide a separate job for each "plugin" (like pipelines, knative, etc.)

Goal:

This epic covers a number of customer requests(RFEs) as well as increases usability.

Why is it important?

Customer satisfaction as well as improved usability.

Acceptance Criteria

  1. Allow user to re-arrange the resources which have ben added to nav by the user
  2. Improved user experience (form based experience)
    1. Form based editing of Routes
    2. Form based creation and editing of Config Maps
    3. Form base creation of Deployments
  3. Improved discovery
    1. Include Share my project on the Add page to increase discoverability
    2. NS Helm Chart Repo
      1. Add tile to Add page for discoverability
      2. Provide a form driven creation experience
      3. User should be able to switch back and forth from Form/YAML
      4. change the intro text to the below & have the link in the intro text bring up the full page form
        1. Browse for charts that help manage complex installations and upgrades. Cluster administrators can customize the content made available in the catalog. Alternatively, developers can try to configure their own custom Helm Chart repository.

Dependencies (External/Internal):

None

Exploration:

Miro board from Epic Exploration

Description

As a user, I want to use a form to create Deployments

Acceptance Criteria

  1. Use existing edit Deployment form component for creating Deployments
  2. Display the form when clicked on `Create Deployment` in the Deployments Search page in the Dev perspective
  3. The `Create Deployment` button in the Deployments list page & the search page in the Admin perspective should have a similar experience.

Additional Details:

Edit deployment form ODC-5007

Description

As a user, I should be able to switch between the form and yaml editor while creating the ProjectHelmChartRepository CR.

Acceptance Criteria

  1. Convert the create form into a form-yaml switcher
  2. Display this form-yaml view in Search -> ProjectHelmChartRepositories in both perspectives

Additional Details:

Form component https://github.com/openshift/console/pull/11227

Problem:

Currently we are only able to get limited telemetry from the Dev Sandbox, but not from any of our managed clusters or on prem clusters.

Goals:

  1. Enable gathering segment telemetry whenever cluster telemetry is enabled on OSD clusters
  2. Have our OSD clusters opt into telemetry by default
  3. Work with PM & UX to identify additional metrics to capture in addition to what we have enabled currently on Sandbox.
  4. Ability to get a single report from woopra across all of our Sandbox and OSD clusters.
  5. Be able to generate a report including metrics of a single cluster or all clusters of a certain type ( sandbox, or OSD)

Why is it important?

In order to improve properly analyze usage and the user experience, we need to be able to gather as much data as possible.

Acceptance Criteria

  1. Extend console backend (bridge) to provide configuration as SERVER_FLAGS
    // JS type
    telemetry?: Record<string, string>
    
    1. Read the annotation of the cluster ConfigMap for telemetry data and pass them into the internal serverconfig.
    2. Pass through this internal serverconfig and export it as SERVER_FLAGS.
    3. Add a new --telemetry CLI option so that the telemetry options could be tested in a dev environment:
      ./bin/bridge --telemetry SEGMENT_API_KEY=a-key-123-xzy
      ./bin/bridge --telemetry CONSOLE_LOG=debug
      
  2. TBD: In best case the new annotation could be read from the cluster ConfigMap...
    1. Otherwise update the console-operator to pass the annotation from the console cluster configuration to the console ConfigMap.

Additional Details:

  1. More information about the integration with the backend could be found in the Telemetry on OSD clusters Google Doc

Goal:
Enhance oc adm release new (and related verbs info, extract, mirror) with heterogeneous architecture support

tl;dr

oc adm release new (and related verbs info, extract, mirror) would be enhanced to optionally allow the creation of manifest list release payloads. The manifest list flow would be triggered whenever the CVO image in an imagestream was a manifest list. If the CVO image is a standard manifest, the generated release payload will also be a manifest. If the CVO image is a manifest list, the generated release payload would be a manifest list (containing a manifest for each arch possessed by the CVO manifest list).

In either case, oc adm release new would permit non-CVO component images to be manifest or manifest lists and pass them through directly to the resultant release manifest(s).

If a manifest list release payload is generated, each architecture specific release payload manifest will reference the same pullspecs provided in the input imagestream.

 

More details in Option 1 of https://docs.google.com/document/d/1BOlPrmPhuGboZbLZWApXszxuJ1eish92NlOeb03XEdE/edit#heading=h.eldc1ppinjjh

Incomplete Epics

This section includes Jira cards that are linked to an Epic, but the Epic itself is not linked to any Feature. These epics were not completed when this image was assembled

Epic Goal

  • Update image registry dependencies (Kubernetes and OpenShift) to the latest versions.

Why is this important?

  • New versions usually bring improvements that are needed by the registry and help with getting updates for z-stream.

Scenarios

  1. As an OpenShift engineer, I want my components to use the versions of dependencies, so that they get fixes for known issues and can be easily updated in z-stream.

Acceptance Criteria

  • CI - MUST be running successfully with tests automated

Dependencies (internal and external)

  1. Kubernetes 1.24

Previous Work (Optional):

  1. IR-210

Done Checklist

  • CI - CI is running, tests are automated and merged.
  • DEV - Upstream code and tests merged: <link to meaningful PR or GitHub Issue>
  • QE - Test plans in Polarion: <link or reference to Polarion>

As a OpenShift engineer
I want image-registry to use the latest k8s libraries
so that image-registry can benefit from new upstream features.

Acceptance criteria

  • image-registry uses k8s.io/api v1.24.z
  • image-registry uses latest openshift/api, openshift/library-go, openshift/client-go

Some of the work done to produce a build for arm64 and to produce custom builds in https://github.com/okd-project/okd-centos9-rebuild required Dockerfiles and similar assets from the cluster operators repositories to be forked. 

 

This story is to track the eventual backport that should be achieved soon to get rid of most of the forks in the repo by merging the "upstream".

Other Complete

This section includes Jira cards that are not linked to either an Epic or a Feature. These tickets were completed when this image was assembled

This is a clone of issue OCPBUGS-17862. The following is the description of the original issue:

This is a clone of issue OCPBUGS-17769. The following is the description of the original issue:

This is a clone of issue OCPBUGS-17568. The following is the description of the original issue:

Description of problem:

 

Customer used Agent-based installer to install 4.13.8 on they CID env, but during install process, the bootstrap machine had oom issue, check sosreport find the init container had oom issue

NOTE: Issue is not see when testing with 4.13.6, per the customer

initContainers:

  • name: machine-config-controller
    image: .Images.MachineConfigOperator
    command: ["/usr/bin/machine-config-controller"]
    args:
  • "bootstrap"
  • "--manifest-dir=/etc/mcc/bootstrap"
  • "--dest-dir=/etc/mcs/bootstrap"
  • "--pull-secret=/etc/mcc/bootstrap/machineconfigcontroller-pull-secret"
  • "--payload-version=.ReleaseVersion"
    resources:
    limits:
    memory: 50Mi

we found the sosreport dmesg and crio logs had oom kill machine-config-controller container issue, the issue was cause by cgroup kill, so looks like the limit 50M is too small

The customer used a physical machine that had 100GB of memory

the customer had some network config in asstant install yaml file, maybe the issue is them had some nic config?

log files:
1. sosreport
https://attachments.access.redhat.com/hydra/rest/cases/03578865/attachments/b5501734-60be-4de4-adcf-da57e22cbb8e?usePresignedUrl=true

2. asstent installer yaml file
https://attachments.access.redhat.com/hydra/rest/cases/03578865/attachments/a32635cf-112d-49ed-828c-4501e95a0e7a?usePresignedUrl=true

3. bootstrap machine oom screenshot
https://attachments.access.redhat.com/hydra/rest/cases/03578865/attachments/eefe2e57-cd23-4abd-9e0b-dd45f20a34d2?usePresignedUrl=true

This is a clone of issue OCPBUGS-4851. The following is the description of the original issue:

This is a clone of issue OCPBUGS-4850. The following is the description of the original issue:

Description of problem:

Kuryr might take a while to create Pods because it has to create Neutron ports for the pods. If a pod gets deleted while this is being processed, a
warning Event will be generated causing the "[sig-network] pods should successfully create sandboxes by adding pod to network" to fail.

Version-Release number of selected component (if applicable):

 

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-18475. The following is the description of the original issue:

This is a clone of issue OCPBUGS-17160. The following is the description of the original issue:

This is a clone of issue OCPBUGS-14049. The following is the description of the original issue:

Description of problem:

After all cluster operators have reconciled after the password rotation, we can still see authentication failures in keystone (attached screenshot of splunk query)

Version-Release number of selected component (if applicable):

Environment:
- OpenShift 4.12.10 on OpenStack 16
- The cluster is managed via RHACM, but password rotation shall be done via "regular"  OpenShift means.

How reproducible:

Rotated the OpenStack credentials according to the documentation [1]

[1] https://docs.openshift.com/container-platform/4.12/authentication/managing_cloud_provider_credentials/cco-mode-passthrough.html#manually-rotating-cloud-creds_cco-mode-passthrough 

Additional info:

- we can't trace back where these authentication failures come from - they do disappear after a cluster upgrade (so when nodes are rebooted and all pods are restarted which indicates that there's still a component using the old credentials)
- The relevant technical integration points _seem_ to be working though (LBaaS, CSI, Machine API, Swift)

What is the business impact? Please also provide timeframe information.

- We cannot rely on splunk monitoring for authentication issues since it's currently constantly showing authentication errors - We cannot be entirely sure that everything works as expected since we don't know the component that doesn't seem to use the new credentials

 

This is a clone of issue OCPBUGS-1428. The following is the description of the original issue:

Description of problem:

When using an OperatorGroup attached to a service account, AND if there is a secret present in the namespace, the operator installation will fail with the message:
the service account does not have any API secret sa=testx-ns/testx-sa
This issue seems similar to https://bugzilla.redhat.com/show_bug.cgi?id=2094303 - which was resolved in 4.11.0 - however, the new element now, is that the presence of a secret in the namespace  is causing the issue.
The name of the secret seems significant - suggesting something somewhere is depending on the order that secrets are listed in. For example, If the secret in the namespace is called "asecret", the problem does not occur. If it is called "zsecret", the problem always occurs.
"zsecret" is not a "kubernetes.io/service-account-token". The issue I have raised here relates to Opaque secrets - zsecret is an Opaque secret. The issue may apply to other types of secrets, but specifically my issue is that when there is an opaque secret present in the namespace, the operator install fails as described. I aught to be allowed to have an opaque secret present in the namespace where I am installing the operator.

Version-Release number of selected component (if applicable):

4.11.0 & 4.11.1

How reproducible:

100% reproducible

Steps to Reproduce:

1.Create namespace: oc new-project testx-ns
2. oc apply -f api-secret-issue.yaml

Actual results:

 

Expected results:

 

Additional info:

API YAML:

cat api-secret-issue.yaml 
apiVersion: v1
kind: Secret
metadata:
  name: zsecret
  namespace: testx-ns
  annotations:
   kubernetes.io/service-account.name: testx-sa
type: Opaque
stringData:
  mykey: mypass

apiVersion: v1
kind: ServiceAccount
metadata:
  name: testx-sa
  namespace: testx-ns

kind: OperatorGroup
apiVersion: operators.coreos.com/v1
metadata:
  name: testx-og
  namespace: testx-ns
spec:
  serviceAccountName: "testx-sa"
  targetNamespaces:
  - testx-ns

apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
  name: testx-role
  namespace: testx-ns
rules:

  • apiGroups: ["*"]
      resources: ["*"]
      verbs: ["*"] 
      

apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
  name: testx-rolebinding
  namespace: testx-ns
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: Role
  name: testx-role
subjects:

  • kind: ServiceAccount
      name: testx-sa
      namespace: testx-ns

apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: etcd-operator
  namespace: testx-ns
spec:
  channel: singlenamespace-alpha
  installPlanApproval: Automatic
  name: etcd
  source: community-operators
  sourceNamespace: openshift-marketplace

This is a clone of issue OCPBUGS-12956. The following is the description of the original issue:

This is a clone of issue OCPBUGS-12910. The following is the description of the original issue:

This is a clone of issue OCPBUGS-12904. The following is the description of the original issue:

Description of problem:

In order to test proxy installations, the CI base image for OpenShift on OpenStack needs netcat.

Description of problem:

Version-Release number of selected component (if applicable):
4.11

How reproducible:
Always

Steps to Reproduce:
1. Enable UWM + dedicated UWM Alertmanager
2. Deploy an application + service monitor + alerting rule which fires always
3. Go to the OCP dev console and silence the alert.

Actual results:
Nothing happens

Expected results:
The alert notification is muted.

Additional info:
Copied from https://bugzilla.redhat.com/show_bug.cgi?id=2100860

Please review the following PR: https://github.com/openshift/images/pull/151

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

Description of problem:

We observed that a dual stack cluster deployed with AI gui only fails.
This cluster is dhcp for ipv4, RA/RS autoconfiguration for ipv6.

It fails with error in the onvkube container

```
I0906 07:45:43.044090   87450 gateway_init.go:261] Initializing Gateway Functionality
I0906 07:45:43.046398   87450 gateway_localnet.go:152] Node local addresses initialized to: map[10.131.31.214:{10.131.31.208 fffffff0} 10.255.0.2:{10.255.0.0 fffffe00} 127.0.0.1:{127.0.0.0 ff000000} 2001:1b74:480:613a:f6e9:d4ff:fef1:6f26:{2001:1b74:480:613a:: ffffffffffffffff0000000000000000} ::1:{::1 ffffffffffffffffffffffffffffffff} fd01:0:0:1::2:{fd01:0:0:1:: ffffffffffffffff0000000000000000} fe80::8ce9:b4ff:fe1a:1208:{fe80:: ffffffffffffffff0000000000000000} fe80::c8ef:ecff:fee3:64c7:{fe80:: ffffffffffffffff0000000000000000} fe80::f6e9:d4ff:fef1:6f26:{fe80:: ffffffffffffffff0000000000000000}]
I0906 07:45:43.047759   87450 helper_linux.go:71] Provided gateway interface "br-ex", found as index: 7
I0906 07:45:43.048045   87450 helper_linux.go:97] Found default gateway interface br-ex 10.131.31.209
I0906 07:45:43.048152   87450 helper_linux.go:71] Provided gateway interface "br-ex", found as index: 7
F0906 07:45:43.048318   87450 ovnkube.go:133] failed to get default gateway interface
```

on the node we observed that there is multi-path entry during

```
default proto ra metric 48 pref medium
        nexthop via fe80::e2f6:2d01:ab14:ec71 dev br-ex weight 1
        nexthop via fe80::e2f6:2d01:ab11:c271 dev br-ex weight 1
```

I manually remove one of the entries (`ip route delete`) and then delete the ovnkube-node pod. Then the installation continues, container works.

Every time there is multiple entry, if the onvkube-node starts, it fails.


Version-Release number of selected component (if applicable):

4.10.30

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

There might a side issue: the interface of the node upon boot takes time to get the ipv6 autoconfiguration, no RS packets seemed to be sent out (observed zero on all routers).

Please review the following PR: https://github.com/openshift/images/pull/156

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

This is a clone of issue OCPBUGS-10497. The following is the description of the original issue:

This is a clone of issue OCPBUGS-10213. The following is the description of the original issue:

This is a clone of issue OCPBUGS-8468. The following is the description of the original issue:

Description of problem:

RHCOS is being published to new AWS regions (https://github.com/openshift/installer/pull/6861) but aws-sdk-go need to be bumped to recognize those regions

Version-Release number of selected component (if applicable):

master/4.14

How reproducible:

always

Steps to Reproduce:

1. openshift-install create install-config
2. Try to select ap-south-2 as a region
3.

Actual results:

New regions are not found. New regions are: ap-south-2, ap-southeast-4, eu-central-2, eu-south-2, me-central-1.

Expected results:

Installer supports and displays the new regions in the Survey

Additional info:

See https://github.com/openshift/installer/blob/master/pkg/asset/installconfig/aws/regions.go#L13-L23

 

Description of problem:

PROXY protocol cannot be enabled for the "Private" endpoint publishing strategy type.

Version-Release number of selected component (if applicable):

4.10.0+. PROXY protocol was made configurable for the "HostNetwork" and "NodePortService" endpoint publishing strategy types, but not for "Private", in this release.

How reproducible:

Always.

Steps to Reproduce:

1. Create an ingresscontroller with the "Private" endpoint publishing strategy type:

oc create -f - <<'EOF'
apiVersion: operator.openshift.io/v1
kind: IngressController
metadata:
  name: example
  namespace: openshift-ingress-operator
spec:
  domain: example.com
  endpointPublishingStrategy:
    type: Private
    private:
      protocol: PROXY
EOF

2. Check the ingresscontroller's status:

oc -n openshift-ingress-operator get ingresscontrollers/example -o 'jsonpath={.status.endpointPublishingStrategy}'

3. Check whether the resulting router deployment has PROXY protocol enabled.

oc -n openshift-ingress get deployments/router-example -o 'jsonpath={.spec.template.spec.containers[0].env[?(@.name=="ROUTER_USE_PROXY_PROTOCOL")]}'

Actual results:

The ingresscontroller is created:

ingresscontroller.operator.openshift.io/example created

The status shows that the spec.endpointPublishingStrategy.private.protocol setting was ignored:

{"private":{},"type":"Private"}

The deployment does not enable PROXY protocol; the oc get command prints no output.

Expected results:

The ingresscontroller's status should indicate that PROXY protocol is enabled:

{"private":{"protocol":"PROXY"},"type":"Private"}

The deployment should have PROXY protocol enabled:

{"name":"ROUTER_USE_PROXY_PROTOCOL","value":"true"}

Additional info:

This bug report duplicates https://bugzilla.redhat.com/show_bug.cgi?id=2104481 in order to facilitate backports.

Since 4.11 OCP comes with OperatorHub definition which declares a capability
and enables all catalog sources. For OKD we want to enable just community-operators
as users may not have Red Hat pull secret set.
This commit would ensure that OKD version of marketplace operator gets
its own OperatorHub manifest with a custom set of operator catalogs enabled

This is a clone of issue OCPBUGS-4489. The following is the description of the original issue:

This is a clone of issue OCPBUGS-4168. The following is the description of the original issue:

Description of problem:

Prometheus continuously restarts due to slow WAL replay

Version-Release number of selected component (if applicable):

openshift - 4.11.13

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-5100. The following is the description of the original issue:

This is a clone of issue OCPBUGS-5068. The following is the description of the original issue:

Description of problem:

virtual media provisioning fails when iLO Ironic driver is used

Version-Release number of selected component (if applicable):

4.13

How reproducible:

Always

Steps to Reproduce:

1. attempt virtual media provisioning on a node configured with ilo-virtualmedia:// drivers
2.
3.

Actual results:

Provisioning fails with "An auth plugin is required to determine endpoint URL" error

Expected results:

Provisioning succeeds

Additional info:

Relevant log snippet:

3742 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector [None req-e58ac1f2-fac6-4d28-be9e-983fa900a19b - - - - - -] Unable to start managed inspection for node e4445d43-3458-4cee-9cbe-6da1de75      78cd: An auth plugin is required to determine endpoint URL: keystoneauth1.exceptions.auth_plugins.MissingAuthPlugin: An auth plugin is required to determine endpoint URL
 3743 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector Traceback (most recent call last):
 3744 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/drivers/modules/inspector.py", line 210, in _start_managed_inspection
 3745 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     task.driver.boot.prepare_ramdisk(task, ramdisk_params=params)
 3746 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic_lib/metrics.py", line 59, in wrapped
 3747 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     result = f(*args, **kwargs)
 3748 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/drivers/modules/ilo/boot.py", line 408, in prepare_ramdisk
 3749 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     iso = image_utils.prepare_deploy_iso(task, ramdisk_params,
 3750 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/drivers/modules/image_utils.py", line 624, in prepare_deploy_iso
 3751 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     return prepare_iso_image(inject_files=inject_files)
 3752 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/drivers/modules/image_utils.py", line 537, in _prepare_iso_image
 3753 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     image_url = img_handler.publish_image(
 3754 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/drivers/modules/image_utils.py", line 193, in publish_image
 3755 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     swift_api = swift.SwiftAPI()
 3756 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector   File "/usr/lib/python3.9/site-packages/ironic/common/swift.py", line 66, in __init__
 3757 2022-12-19T19:02:05.997747170Z 2022-12-19 19:02:05.995 1 ERROR ironic.drivers.modules.inspector     endpoint = keystone.get_endpoint('swift', session=session)

This is a clone of issue OCPBUGS-7409. The following is the description of the original issue:

This is a clone of issue OCPBUGS-7374. The following is the description of the original issue:

Originally reported by lance5890 in issue https://github.com/openshift/cluster-etcd-operator/issues/1000

The controllers sometimes get stuck on listing members in failure scenarios, this is known and can be mitigated by simply restarting the CEO. 

similar BZ 2093819 with stuck controllers was fixed slightly different in https://github.com/openshift/cluster-etcd-operator/commit/4816fab709e11e0681b760003be3f1de12c9c103

 

This fix was contributed by lance5890, thanks a lot!

 

Following the trail
https://github.com/kubernetes-sigs/aws-ebs-csi-driver/issues/1139
https://github.com/kubernetes-sigs/aws-ebs-csi-driver/pull/1175
https://github.com/openshift/aws-ebs-csi-driver/pull/206
 
Looks like the fix should be in 4.12, but it still see it being 39 vs ~24 on an m6i instance type.

It seems that the kubelet applies this capacity to the node in 4.11 and earlier and, thus, unlikely to receive this fix for attachable volumes in the upstream CSI driver. 4.12 behavior is currently unknown but it seems that the kubelet might still be setting this capacity.

The actual issue is that kube scheduler schedules pods that require PVs to nodes where those PVs can not be attached.

This is a clone of issue OCPBUGS-6517. The following is the description of the original issue:

Description of problem:

When the cluster is configured with Proxy the swift client in the image registry operator is not using the proxy to authenticate with OpenStack, so it's unable to reach the OpenStack API. This issue became evident since recently the support was added to not fallback to cinder in case swift is available[1].

[1]https://github.com/openshift/cluster-image-registry-operator/pull/819

 

Version-Release number of selected component (if applicable):

 

How reproducible:

 

Steps to Reproduce:

1. Deploy a cluster with proxy and restricted installation
2. 
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-6661. The following is the description of the original issue: 

Description of problem:

CRL list is capped at 1MB due to configmap max size. If multiple public CRLs are needed for ingress controller the CRL pem file will be over 1MB. 

Version-Release number of selected component (if applicable):

 

How reproducible:

100%

Steps to Reproduce:

1. Create CRL configmap with the following distribution points: 

         Issuer: C=US, O=DigiCert Inc, CN=DigiCert Global G2 TLS RSA SHA256 2020 CA1
         Subject: SOME SIGNED CERT            X509v3 CRL Distribution Points: 
                Full Name:
                  URI:http://crl3.digicert.com/DigiCertGlobalG2TLSRSASHA2562020CA1-2.cr  
       
      
# curl -o DigiCertGlobalG2TLSRSASHA2562020CA1-2.crl http://crl3.digicert.com/DigiCertGlobalG2TLSRSASHA2562020CA1-2.crl
# openssl crl -in  DigiCertGlobalG2TLSRSASHA2562020CA1-2.crl -inform DER -out  DigiCertGlobalG2TLSRSASHA2562020CA1-2.pem 
# du -bsh DigiCertGlobalG2TLSRSASHA2562020CA1-2.pem 
604K    DigiCertGlobalG2TLSRSASHA2562020CA1-2.pem


I still need to find more intermediate CRLS to grow this. 

Actual results:

2023-01-25T13:45:01.443Z ERROR operator.init controller/controller.go:273 Reconciler error {"controller": "crl", "object": {"name":"custom","namespace":"openshift-ingress-operator"}, "namespace": "openshift-ingress-operator", "name": "custom", "reconcileID": "d49d9b96-d509-4562-b3d9-d4fc315226c0", "error": "failed to ensure client CA CRL configmap for ingresscontroller openshift-ingress-operator/custom: failed to update configmap: ConfigMap \"router-client-ca-crl-custom\" is invalid: []: Too long: must have at most 1048576 bytes"}

Expected results:

First be able to create a configmap where data only accounted to the 1MB max (see additional info below for more details), second some way to compress or allow a large CRL list that would be larger than 1MB

Additional info:

Only using this CRL and it being only 600K still causes issue and it could be due to  the `last-applied-configuration` annotation on the configmap. This is added since we do an apply operation (update) on the configmap. I am not sure if this is counting towards the 1MB max. 

https://github.com/openshift/cluster-ingress-operator/blob/release-4.10/pkg/operator/controller/crl/crl_configmap.go#L295 

Not sure if we could just replace the configmap.   

 

Description of problem:

WebhookConfiguration caBundle injection is incorrect when some webhooks already configured with caBundle.

Behavior seems to be that the first n number of webhooks in `.webhooks` array have caBundle injected, where n is the number of webhooks that do not have caBundle set.

Version-Release number of selected component (if applicable):

 

How reproducible

 

Steps to Reproduce:

1. Create a validatingwebhookconfigurations or mutatingwebhookconfigurations with `service.beta.openshift.io/inject-cabundle: "true"` annotation.

2. oc edit validatingwebhookconfigurations (or oc edit mutatingwebhookconfigurations)

3. Add a new webhook to the end of the list `.webhooks`. It will not have caBundle set manually as service-ca should inject it. 

4. Observe new webhook does not get caBundle injected.

Note: it is important in step. 3 that the new webhook is added to the end of the list. 

 

Actual results:

Only the first n webhooks have caBundle injected where n is the number of webhooks without caBundle set.

Expected results:

All webhooks have caBundle injected when they do not have it set.

Additional info:

Open PR here: https://github.com/openshift/service-ca-operator/pull/207

The issue seems to be a mistake with go-lang for range syntax where "i" is the index of desired "i" to update.  

tl dr; code should update the value of the int in the array, not the index of the int in the array.

Please review the following PR: https://github.com/openshift/images/pull/159

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

Description of problem:

The alertmanager pod is stuck on OCP 4.11 with OVN in container Creating State

From oc describe alertmanager pod:
...
Events:
  Type     Reason                  Age                  From     Message
  ----     ------                  ----                 ----     -------
  Warning  FailedCreatePodSandBox  16s (x459 over 17h)  kubelet  (combined from similar events): Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_alertmanager-managed-ocs-alertmanager-0_openshift-storage_3a55ed54-4eaa-4f65-8a10-e5d21fad1ebc_0(88575547dc0b210307b89dd2bb8e379ece0962b607ac2707a1c2cf630b1aaa78): error adding pod openshift-storage_alertmanager-managed-ocs-alertmanager-0 to CNI network "multus-cni-network": plugin type="multus" name="multus-cni-network" failed (add): [openshift-storage/alertmanager-managed-ocs-alertmanager-0/3a55ed54-4eaa-4f65-8a10-e5d21fad1ebc:ovn-kubernetes]: error adding container to network "ovn-kubernetes": CNI request failed with status 400: '[openshift-storage/alertmanager-managed-ocs-alertmanager-0 88575547dc0b210307b89dd2bb8e379ece0962b607ac2707a1c2cf630b1aaa78] [openshift

Version-Release number of selected component (if applicable):

OCP 4.11 with OVN

How reproducible:

100%

Steps to Reproduce:

1. Terminate the node on which alertmanager pod is running
2. pod will get stuck in container Creating state
3.

Actual results:

AlertManager pod is stuck in container Creating state

Expected results:

Alertmanager pod is ready

Additional info:

The workaround would be to terminate the alertmanager pod

This is a clone of issue OCPBUGS-8261. The following is the description of the original issue:

Description of problem:

An update from 4.13.0-ec.2 to 4.13.0-ec.3 stuck on:

$ oc get clusteroperator machine-config
NAME             VERSION       AVAILABLE   PROGRESSING   DEGRADED   SINCE   MESSAGE
machine-config   4.13.0-ec.2   True        True          True       30h     Unable to apply 4.13.0-ec.3: error during syncRequiredMachineConfigPools: [timed out waiting for the condition, error pool worker is not ready, retrying. Status: (pool degraded: true total: 105, ready 105, updated: 105, unavailable: 0)]

The worker MachineConfigPool status included:

Unable to find source-code formatter for language: node. Available languages are: actionscript, ada, applescript, bash, c, c#, c++, cpp, css, erlang, go, groovy, haskell, html, java, javascript, js, json, lua, none, nyan, objc, perl, php, python, r, rainbow, ruby, scala, sh, sql, swift, visualbasic, xml, yaml      type: NodeDegraded
    - lastTransitionTime: "2023-02-16T14:29:21Z"
      message: 'Failed to render configuration for pool worker: Ignoring MC 99-worker-generated-containerruntime
        generated by older version 8276d9c1f574481043d3661a1ace1f36cd8c3b62 (my version:
        c06601510c0917a48912cc2dda095d8414cc5182)'

Version-Release number of selected component (if applicable):

4.13.0-ec.3. The behavior was apparently introduced as part of OCPBUGS-6018, which has been backported, so the following update targets are expected to be vulnerable: 4.10.52+, 4.11.26+, 4.12.2+, and 4.13.0-ec.3.

How reproducible:

100%, when updating into a vulnerable release, if you happen to have leaked MachineConfig.

Steps to Reproduce:

1. 4.12.0-ec.1 dropped cleanUpDuplicatedMC. Run a later release, like 4.13.0-ec.2.
2. Create more than one KubeletConfig or ContainerRuntimeConfig targeting the worker pool (or any pool other than master). The number of clusters who have had redundant configuration objects like this is expected to be small.
3. (Optionally?) delete the extra KubeletConfig and ContainerRuntimeConfig.
4. Update to 4.13.0-ec.3.

Actual results:

Update sticks on the machine-config ClusterOperator, as described above.

Expected results:

Update completes without issues.

Description of problem:

Sometimes we see VMs fail to power on when the land on a host that does not have enough resources. The current power on does not retry or leverage DRS to power on the node on a suitable host.

https://github.com/vmware/govmomi/issues/1026

Our code is still making calls to PowerOnVM_Task which, according to the vsphere docs, is deprecated and we should use PowerOnMultiVM_Task instead.

PowerOnVM_Task does not return a DRS ClusterRecommendation, no vmotion nor host power operations will be done as part of a DRS-facilitated power on. To have DRS consider such operations use PowerOnMultiVM_Task.

https://vdc-download.vmware.com/vmwb-repository/dcr-public/b50dcbbf-051d-4204-a3e7-e1b618c1e384/538cf2ec-b34f-4bae-a332-3820ef9e7773/vim.VirtualMachine.html#powerOn:
As of vSphere API 5.1, use of this method with vCenter Server is deprecated; use PowerOnMultiVM_Task instead.

Version-Release number of selected component (if applicable):
4.8.x

How reproducible:
Always

Steps to Reproduce:
1.
2.
3.

Actual results:
Sometimes powers on fails requiring manual intervention.

Expected results:
PowerOn should use DRS to ensure it's always successful.

Additional info:

Please review the following PR: https://github.com/openshift/images/pull/133

The PR has been automatically opened by ART (#aos-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #aos-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

Please review the following PR: https://github.com/openshift/service-ca-operator/pull/221

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

This is a clone of issue OCPBUGS-723. The following is the description of the original issue:

Description of problem:
I have a customer who created clusterquota for one of the namespace, it got created but the values were not reflecting under limits or not displaying namespace details.
~~~
$ oc describe AppliedClusterResourceQuota
Name: test-clusterquota
Created: 19 minutes ago
Labels: size=custom
Annotations: <none>
Namespace Selector: []
Label Selector:
AnnotationSelector: map[openshift.io/requester:system:serviceaccount:application-service-accounts:test-sa]
Scopes: NotTerminating
Resource Used Hard
-------- ---- ----
~~~

WORKAROUND: They recreated the clusterquota object (cache it off, delete it, create new) after which it displayed values as expected.

In the past, they saw similar behavior on their test cluster, there it was heavily utilized the etcd DB was much larger in size (>2.5Gi), and had many more objects (at that time, helm secrets were being cached for all deployments, and keeping a history of 10, so etcd was being bombarded).

This cluster the same "symptom" was noticed however etcd was nowhere near that in size nor the amount of etcd objects and/or helm cached secrets.

Version-Release number of selected component (if applicable): OCP 4.9

How reproducible: Occurred only twice(once in test and in current cluster)

Steps to Reproduce:
1. Create ClusterQuota
2. Check AppliedClusterResourceQuota
3. The values and namespace is empty

Actual results: ClusterQuota should display the values

Expected results: ClusterQuota not displaying values

This is a clone of issue OCPBUGS-15643. The following is the description of the original issue:

This is a clone of issue OCPBUGS-15606. The following is the description of the original issue:

This is a clone of issue OCPBUGS-15497. The following is the description of the original issue:

I am using a BuildConfig with git source and the Docker strategy. The git repo contains a large zip file via LFS and that zip file is not getting downloaded. Instead just the ascii metadata is getting downloaded. I've created a simple reproducer (https://github.com/selrahal/buildconfig-git-lfs) on my personal github. If you clone the repo

git clone git@github.com:selrahal/buildconfig-git-lfs.git

and apply the bc.yaml file with

oc apply -f bc.yaml

Then start the build with

oc start-build test-git-lfs

You will see the build fails at the unzip step in the docker file

STEP 3/7: RUN unzip migrationtoolkit-mta-cli-5.3.0-offline.zip
End-of-central-directory signature not found. Either this file is not
a zipfile, or it constitutes one disk of a multi-part archive. In the
latter case the central directory and zipfile comment will be found on
the last disk(s) of this archive.

I've attached the full build logs to this issue.

This is a clone of issue OCPBUGS-20100. The following is the description of the original issue:

This is a clone of issue OCPBUGS-19378. The following is the description of the original issue:

Description of problem:

In 4.14 we have introduced a bunch of improvements to the backup/restore scripts along with many test cases and a full e2e suite. 

Since we've already rebased the new features for etcdctl into all supported branches, we can also backport the CEO tooling along with it to get into a consistent and supportable state again. 

Version-Release number of selected component (if applicable):

4.11->4.13

How reproducible:

na

Steps to Reproduce:

na

Actual results:

na

Expected results:

na

Additional info:

 

Clone of https://bugzilla.redhat.com/show_bug.cgi?id=2106803 to backport the e2e fix to 4.11 and 4.10.

Description of problem: E2E: intermittent failure is seen on tests for devfile due to network call to devfile registry

Deploy git workload with devfile from topology page: A-04-TC01

https://gcsweb-ci.apps.ci.l2s4.p1.openshiftapps.com/gcs/origin-ci-test/pr-logs/pull/openshift_console/11726/pull-ci-openshift-console-master-e2e-gcp-console/1547046629775773696/artifacts/e2e-gcp-console/test/artifacts/gui_test_screenshots/cypress/screenshots/e2e/add-flow-ci.feature/Deploy%20git%20workload%20with%20devfile%20from%20topology%20page%20A-04-TC01%20--%20after%20each%20hook%20(failed).png

https://prow.ci.openshift.org/view/gs/origin-ci-test/pr-logs/pull/openshift_console/11768/pull-ci-openshift-console-master-e2e-gcp-console/1547061730478133248

Version-Release number of selected component (if applicable):

How reproducible: Intermittent

Steps to Reproduce:
1. Run test for add-flow-ci.feature to test Deploy git workload with devfile from topology page: A-04-TC01

Actual results:

Expected results: Show always pass

Additional info:

Please review the following PR: https://github.com/openshift/images/pull/183

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

This is a clone of issue OCPBUGS-7474. The following is the description of the original issue:

This is a clone of issue OCPBUGS-6714. The following is the description of the original issue:

Description of problem:

Traffic from egress IPs was interrupted after Cluster patch to Openshift 4.10.46

a customer cluster was patched. It is an Openshift 4.10.46 cluster with SDN.

More description about issue is available in private comment below since it contains customer data.

This is a clone of issue OCPBUGS-1805. The following is the description of the original issue:

The vSphere CSI cloud.conf lists the single datacenter from platform workspace config but in a multi-zone setup (https://github.com/openshift/enhancements/pull/918 ) there may be more than the one datacenter.

This issue is resulting in PVs failing to attach because the virtual machines can't be find in any other datacenter. For example:

0s Warning FailedAttachVolume pod/image-registry-85b5d5db54-m78vp AttachVolume.Attach failed for volume "pvc-ab1a0611-cb3b-418d-bb3b-1e7bbe2a69ed" : rpc error: code = Internal desc = failed to find VirtualMachine for node:"rbost-zonal-ghxp2-worker-3-xm7gw". Error: virtual machine wasn't found  

The machine above lives in datacenter-2 but the CSI cloud.conf is only aware of the datacenter IBMCloud.

$ oc get cm vsphere-csi-config -o yaml  -n openshift-cluster-csi-drivers | grep datacenters
    datacenters = "IBMCloud" 

 

This is a clone of issue OCPBUGS-6755. The following is the description of the original issue:

This is a clone of issue OCPBUGS-3316. The following is the description of the original issue:

Description of problem:

Branch name in repository pipelineruns list view should match the actual github branch name.

Version-Release number of selected component (if applicable):

4.11.z

How reproducible:

alwaus

Steps to Reproduce:

1. Create a repository
2. Trigger the pipelineruns by push or pull request event on the github 

Actual results:

Branch name contains "refs-heads-" prefix in front of the actual branch name eg: "refs-heads-cicd-demo" (cicd-demo is the branch name)

Expected results:

Branch name should be the acutal github branch name. just `cicd-demo`should be shown in the branch column.

 

Additional info:
Ref: https://coreos.slack.com/archives/CHG0KRB7G/p1667564311865459

Description of problem:

Customer is facing issue similar to https://github.com/devfile/api/issues/897

Version-Release number of selected component (if applicable):

OCP 4.10.17

How reproducible:
N/A
Steps to Reproduce:
1.
2.
3.

Actual results:

Expected results:

Additional info:

Tried working around it with ALL_PROXY but it did not help. Note because the console operator reverts changes pretty quickly testing this was a bit of a PITA

This is a clone of issue OCPBUGS-10225. The following is the description of the original issue:

Description of problem:
Pipeline Repository (Pipeline-as-code) list never shows an Event type.

Version-Release number of selected component (if applicable):
4.9+

How reproducible:
Always

Steps to Reproduce:

  1. Install Pipelines Operator and setup a Pipeline-as-code repository
  2. Trigger an event and a build

Actual results:
Pipeline Repository list shows a column Event type but no value.

Expected results:
Pipeline Repository list should show the Event type from the matching Pipeline Run.

Similar to the Pipeline Run Details page based on the label.

Additional info:
The list page packages/pipelines-plugin/src/components/repository/list-page/RepositoryRow.tsx renders obj.metadata.namespace as event type.

I believe we should show the Pipeline Run event type instead. packages/pipelines-plugin/src/components/repository/RepositoryLinkList.tsx uses

{plrLabels[RepositoryLabels[RepositoryFields.EVENT_TYPE]]}

to render it.

Also the Pipeline Repository details page tried to render the Branch and Event type from the Repository resource. My research says these properties doesn't exist on the Repository resource. The code should be removed from the Repository details page.

This is a clone of issue OCPBUGS-13811. The following is the description of the original issue:

This is a clone of issue OCPBUGS-10816. The following is the description of the original issue:

Description of problem:

We have observed a situation where:
- A workload mounting multiple EBS volumes gets stuck in a Terminating state when it finishes.
- The node that the workload ran on eventually gets stuck draining, because it gets stuck on unmounting one of the volumes from that workload, despite no containers from the workload now running on the node.

What we observe via the node logs is that the volume seems to unmount successfully. Then it attempts to unmount a second time, unsuccessfully. This unmount attempt then repeats and holds up the node.

Specific examples from the node's logs to illustrate this will be included in a private comment. 

Version-Release number of selected component (if applicable):

4.11.5

How reproducible:

Has occurred on four separate nodes on one specific cluster, but the mechanism to reproduce it is not known.

Steps to Reproduce:

1.
2.
3.

Actual results:

A volume gets stuck unmounting, holding up removal of the node and completed deletion of the pod.

Expected results:

The volume should not get stuck unmounting.

Additional info:

 

This is a clone of issue OCPBUGS-3265. The following is the description of the original issue:

This is a clone of issue OCPBUGS-3172. The following is the description of the original issue:

Customer is trying to install the Logging operator, which appears to attempt to install a dynamic plugin. The operator installation fails in the console because permissions aren't available to "patch resource consoles".

We shouldn't block operator installation if permission issues prevent dynamic plugin installation.

This is an OSD cluster, presumably for a customer with "cluster-admin", although it may be a paired down permission set called "dedicated-admin".

See https://docs.google.com/document/d/1hYS-bm6aH7S6z7We76dn9XOFcpi9CGYcGoJys514YSY/edit for permissions investigation work on OSD

This is a clone of issue OCPBUGS-5067. The following is the description of the original issue:

Description of problem:

Since coreos-installer writes to stdout, its logs are not available for us.

Version-Release number of selected component (if applicable):

 

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-6764. The following is the description of the original issue:

Description of problem:
The "Add Git Repository" has a "Show configuration options" expandable section that shows the required permissions for a webhook setup, and provides a link to "read more about setting up webhook".

But the permission section shows nothing when open this second expandable section, and the link doesn't do anything until the user enters a "supported" GitHub, GitLab or BitBucket URL.

Version-Release number of selected component (if applicable):
4.11-4.13

How reproducible:
Always

Steps to Reproduce:

  1. Install Pipelines operator
  2. Navigate to the Developer perspective > Pipelines
  3. Press "Create" and select "Repository"
  4. Click on "Show configuration options"
  5. Click on "See Git permissions"
  6. Click on "Read more about setting up webhook"

Actual results:

  1. The Git permission section shows no git permissions.
  2. The Read more link doesn't open any new page.

Expected results:

  1. The Git permission section should show some info or must not be disabled.
  2. The Read more link should open a page or must not be displayed as well.

Additional info:

  1. None

This is a clone of issue OCPBUGS-6816. The following is the description of the original issue:

This is a clone of issue OCPBUGS-6799. The following is the description of the original issue:

Description of problem:
The pipelines -> repositories list view in Dev Console does not show the running pipelineline as the last pipelinerun in the table.

Original BugZilla Link: https://bugzilla.redhat.com/show_bug.cgi?id=2016006
OCPBUGSM: https://issues.redhat.com/browse/OCPBUGSM-36408

Created attachment 1905034 [details]
Plugin page with error

Steps to reproduce:

1. Install a plugin with a page that has a runtime error. (Demo Plugin -> Dynamic Nav 1 currently has an error for me, but you can reproduce by editing any plugin and introducing an error.)
2. Observe the "something went wrong" error message.
3. Navigate to any other page (e.g. Workloads -> Pods)

Expected result:

The pods page is displayed.

Action result:

The error message persists. There is no way to clear except to refresh the browser.

This is a clone of issue OCPBUGS-16151. The following is the description of the original issue:

This is a clone of issue OCPBUGS-13812. The following is the description of the original issue:

This is a clone of issue OCPBUGS-13718. The following is the description of the original issue:

Description of problem:

IPI install on azure stack failed when setting platform.azure.osDiks.diskType as StandardSSD_LRS in install-config.yaml.

When setting controlPlane.platform.azure.osDisk.diskType as StandardSSD_LRS, get error in terraform log and some resources have been created.

level=error msg=Error: expected storage_os_disk.0.managed_disk_type to be one of [Premium_LRS Standard_LRS], got StandardSSD_LRS
level=error
level=error msg=  with azurestack_virtual_machine.bootstrap,
level=error msg=  on main.tf line 107, in resource "azurestack_virtual_machine" "bootstrap":
level=error msg= 107: resource "azurestack_virtual_machine" "bootstrap" {
level=error
level=error msg=failed to fetch Cluster: failed to generate asset "Cluster": failure applying terraform for "bootstrap" stage: failed to create cluster: failed to apply Terraform: exit status 1
level=error
level=error msg=Error: expected storage_os_disk.0.managed_disk_type to be one of [Premium_LRS Standard_LRS], got StandardSSD_LRS
level=error
level=error msg=  with azurestack_virtual_machine.bootstrap,
level=error msg=  on main.tf line 107, in resource "azurestack_virtual_machine" "bootstrap":
level=error msg= 107: resource "azurestack_virtual_machine" "bootstrap" {
level=error
level=error

When setting compute.platform.azure.osDisk.diskType as StandardSSD_LRS, fail to provision compute machines

$ oc get machine -n openshift-machine-api
NAME                                     PHASE     TYPE              REGION   ZONE   AGE
jima414ash03-xkq5x-master-0              Running   Standard_DS4_v2   mtcazs          62m
jima414ash03-xkq5x-master-1              Running   Standard_DS4_v2   mtcazs          62m
jima414ash03-xkq5x-master-2              Running   Standard_DS4_v2   mtcazs          62m
jima414ash03-xkq5x-worker-mtcazs-89mgn   Failed                                      52m
jima414ash03-xkq5x-worker-mtcazs-jl5kk   Failed                                      52m
jima414ash03-xkq5x-worker-mtcazs-p5kvw   Failed                                      52m

$ oc describe machine jima414ash03-xkq5x-worker-mtcazs-jl5kk -n openshift-machine-api
...
  Error Message:           failed to reconcile machine "jima414ash03-xkq5x-worker-mtcazs-jl5kk": failed to create vm jima414ash03-xkq5x-worker-mtcazs-jl5kk: failure sending request for machine jima414ash03-xkq5x-worker-mtcazs-jl5kk: cannot create vm: compute.VirtualMachinesClient#CreateOrUpdate: Failure sending request: StatusCode=400 -- Original Error: Code="InvalidParameter" Message="Storage account type 'StandardSSD_LRS' is supported by Microsoft.Compute API version 2018-04-01 and above" Target="osDisk.managedDisk.storageAccountType"
...

Based on azure-stack doc[1], supported disk types on ASH are Premium SSD, Standard HDD. It's better to do validation for diskType on Azure Stack to avoid above errors.

[1]https://learn.microsoft.com/en-us/azure-stack/user/azure-stack-managed-disk-considerations?view=azs-2206&tabs=az1%2Caz2#cheat-sheet-managed-disk-differences

Version-Release number of selected component (if applicable):

4.14.0-0.nightly-2023-05-16-085836

How reproducible:

Always

Steps to Reproduce:

1. Prepare install-config.yaml, set platform.azure.osDiks.diskType as StandardSSD_LRS
2. Install IPI cluster on Azure Stack
3.

Actual results:

Installation failed

Expected results:

Installer validate diskType on AzureStack Cloud, and exit for unsupported disk type with error message

Additional info:

 

This is a clone of issue OCPBUGS-4311. The following is the description of the original issue:

This is a clone of issue OCPBUGS-4305. The following is the description of the original issue:

Description of problem:

Please add an option to DISABLE debug in ironic-api. Presently it is enabled by default and there is no way to disable it or reduce log level

https://github.com/metal3-io/ironic-image/blob/main/ironic-config/ironic.conf.j2#L3


Version-Release number of selected component (if applicable): none

How reproducible: Every time

Steps to Reproduce:

Please check source code here: https://github.com/metal3-io/ironic-image/blob/main/ironic-config/ironic.conf.j2#L3

It is enabled by default and there is no way to disable it or reduce log level

Actual results:

Please check Case: 03371411, the log file grew to 409 GB

Expected results: Need a way to disable debug

Additional info: Case 03371411. A cluster must gather and log file can be found in the case.

This is a clone of issue OCPBUGS-7732. The following is the description of the original issue:

Description of problem:

When services are deleted, the services controller cache should also remove the service from its top level cache to avoid growing forever.

While this is not an issue in 4.13 once the lb_cache rework merges [1], the 4.12 and older branches have this problem because that rework is meant for 4.13 only.

[1]: https://github.com/ovn-org/ovn-kubernetes/pull/3387

This is the location where alreadyApplied is not deleting the removal: 
https://github.com/openshift/ovn-kubernetes/blob/cf9fb51510e1870961bf3a0f064b73536757a4f8/go-controller/pkg/ovn/controller/services/services_controller.go#L269

It should do the similar changes depicted here (currently merged upstream):
https://github.com/ovn-org/ovn-kubernetes/blob/cd78ae1af4657d38bdc41003a8737aa958d62b9d/go-controller/pkg/ovn/controller/services/services_controller.go#L322-L324

 

Version-Release number of selected component (if applicable):

 

How reproducible:

100%

Steps to Reproduce:

1. create service -- use unique name
2. remove service
3. notice how alreadyApplied grows and never gets smaller
4. repeat

Actual results:

^^

Expected results:

alreadyApplied should not grow forever

Additional info:

 

This is a clone of issue OCPBUGS-18305. The following is the description of the original issue:

Description of problem:

It appears it may be possible to have invalid CSV entries in the resolver cache, resulting in the inability to reinstall an Operator.

The situation:
--------------
A customer has removed the CSV, InstallPlan and Subscription for the GitOps Operator from the cluster but upon attempting to reinstall the Operator, the OLM was providing a conflict with existing CSV.

This CSV was not in the ETCD instance and was removed previously. Upon deleting the `operator-catalog` and `operator-lifecycle-manager` Pods, the collision was resolved and the Operator was able to installed again.
~~~
'Warning' reason: 'ResolutionFailed' constraints not satisfiable: subscription openshift-gitops-operator exists, subscription openshift-gitops-operator requires redhat-operators/openshift-marketplace/stable/openshift-gitops-operator.v1.5.8, redhat-operators/openshift-marketplace/stable/openshift-gitops-operator.v1.5.8 and @existing/openshift-operators//openshift-gitops-operator.v1.5.6-0.1664915551.p originate from package openshift-gitops-operator, clusterserviceversion openshift-gitops-operator.v1.5.6-0.1664915551.p exists and is not referenced by a subscription
~~~

Version-Release number of selected component (if applicable):

4.9.31

How reproducible:

Very intermittent, however once the issue has occurred it was impossible to avoid without deleting the Pods.

Steps to Reproduce:

1. Add Operator with manual approval InstallPlan
2. Remove Operator (Subscription, CSV, InstallPlan)
3. Attempt to reinstall Operator 

Actual results:

Very intermittent failure

Expected results:

Operators do not have conflicts with CSVs which have already been removed.

Additional info:

Briefly reviewing the OLM code, it appears an internal resolver cache is populated and used for checking constraints when an operator is installed. If there are stale entries in the cache, this would result in the described issue.
The cache appears to have been rearchitected (moved to a dedicated object) since OCP 4.9.31. Due to the nature of this issue, the request does not have clear reproduction steps to reproduce so if the issue is unable to reproduced, I would like instructions on how to dump the contents of the cache if the issue is to arise again.

This is a clone of issue OCPBUGS-4072. The following is the description of the original issue:

This is a clone of issue OCPBUGS-4026. The following is the description of the original issue:

Description of problem:
There is an endless re-render loop and a browser feels slow to stuck when opening the add page or the topology.

Saw also endless API calls to /api/kubernetes/apis/binding.operators.coreos.com/v1alpha1/bindablekinds/bindable-kinds

Version-Release number of selected component (if applicable):
1. Console UI 4.12-4.13 (master)
2. Service Binding Operator (tested with 1.3.1)

How reproducible:
Always with installed SBO

But the "stuck feeling" depends on the browser (Firefox feels more stuck) and your locale machine power

Steps to Reproduce:
1. Install Service Binding Operator
2. Create or update the BindableKinds resource "bindable-kinds"

apiVersion: binding.operators.coreos.com/v1alpha1
kind: BindableKinds
metadata:
  name: bindable-kinds

3. Open the browser console log
4. Open the console UI and navigate to the add page

Actual results:
1. Saw endless API calls to /api/kubernetes/apis/binding.operators.coreos.com/v1alpha1/bindablekinds/bindable-kinds
2. Browser feels slow and get stuck after some time
3. The page crashs after some time

Expected results:
1. The API call should be called just once
2. The add page should just work without feeling laggy
3. No crash

Additional info:
Get introduced after we watching the bindable-kinds resource with https://github.com/openshift/console/pull/11161

It looks like this happen only if the SBO is installed and the bindable-kinds resource exist, but doesn't contain any status.

The status list all available bindable resource types. I could not reproduce this by installing and uninstalling an operator, but you can manually create or update this resource as mentioned above.

Description of problem: This is a follow-up to OCPBUGS-2795 and OCPBUGS-2941.

The installer fails to destroy the cluster when the OpenStack object storage omits 'content-type' from responses. This can happen on responses with HTTP status code 204, where a reverse proxy is truncating content-related headers (see this nginX bug report). In such cases, the Installer errors with:

level=error msg=Bulk deleting of container "5ifivltb-ac890-chr5h-image-registry-fnxlmmhiesrfvpuxlxqnkoxdbl" objects failed: Cannot extract names from response with content-type: []

Listing container object suffers from the same issue as listing the containers and this one isn't fixed in latest versions of gophercloud. I've reported https://github.com/gophercloud/gophercloud/issues/2509 and fixing it with https://github.com/gophercloud/gophercloud/issues/2510, however we likely won't be able to backport the bump to gophercloud master back to release-4.8 so we'll have to look for alternatives.

I'm setting the priority to critical as it's causing all our jobs to fail in master.

Version-Release number of selected component (if applicable):

4.8.z

How reproducible:

Likely not happening in customer environments where Swift is exposed directly. We're seeing the issue in our CI where we're using a non-RHOSP managed cloud.

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

Description of problem:

    

Version-Release number of selected component (if applicable):

    

How reproducible:

    

Steps to Reproduce:

    1.
    2.
    3.
    

Actual results:

    

Expected results:

    

Additional info:

    

Since HyperShift / Hosted Control Plane have adopted include.release.openshift.io/ibm-cloud-managed, to tailor the resources of clusters running in the ROKS IBM environment, the include.release.openshift.io/hypershift addition will allow Hypershift to express different profile choices than ROKS

Description of problem:

During restart egress firewall acls will be deleted and re-created from scratch, meaning that egress firewall rules won't be applied for some time during restart

Version-Release number of selected component (if applicable):

 

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

Description of problem:
Switching the spec.endpointPublishingStrategy.loadBalancer.scope of the default ingresscontroller results in a degraded ingress operator. The routes using that endpoint like the console URL become inaccessible.
Degraded operators after scope change:

$ oc get co | grep -v ' True        False         False'
NAME                                       VERSION   AVAILABLE   PROGRESSING   DEGRADED   SINCE   MESSAGE
authentication                             4.11.4    False       False         True       72m     OAuthServerRouteEndpointAccessibleControllerAvailable: Get "https://oauth-openshift.apps.kartrosa.ukld.s1.devshift.org/healthz": EOF
console                                    4.11.4    False       False         False      72m     RouteHealthAvailable: failed to GET route (https://console-openshift-console.apps.kartrosa.ukld.s1.devshift.org): Get "https://console-openshift-console.apps.kartrosa.ukld.s1.devshift.org": EOF
ingress                                    4.11.4    True        False         True       65m     The "default" ingress controller reports Degraded=True: DegradedConditions: One or more other status conditions indicate a degraded state: CanaryChecksSucceeding=False (CanaryChecksRepetitiveFailures: Canary route checks for the default ingress controller are failing)

We have noticed that each time this happens the underlying AWS loadbalancer gets recreated which is as expected however the router pods probably do not get notified about the new loadbalancer. The instances in the new loadbalancer become 'outOfService'.

Restarting one of the router pods fixes the issue and brings back a couple of instances under the loadbalancer back to 'InService' which leads to the operators becoming happy again.

Version-Release number of selected component (if applicable):

ingress in 4.11.z however we suspect this issue to also apply to older versions

How reproducible:

Consistently reproducible

Steps to Reproduce:

1. Create a test OCP 4.11 cluster in AWS
2. Switch the spec.endpointPublishingStrategy.loadBalancer.scope of the default ingresscontroller in openshift-ingress-operator to Internal from External (or vice versa)
3. New Loadbalancer is created in AWS for the default router service, however the instances behind are not in service

Actual results:

ingress, authentication and console operators go into a degraded state. Console URL of the cluster is inaccessible

Expected results:

The ingresscontroller scope transition from internal->External (or vice versa) is smooth without any downtime or operators going into degraded state. The console is accessible.

 

Description of problem:

Similar to OCPBUGS-11636 ccoctl needs to be updated to account for the s3 bucket changes described in https://aws.amazon.com/blogs/aws/heads-up-amazon-s3-security-changes-are-coming-in-april-of-2023/

these changes have rolled out to us-east-2 and China regions as of today and will roll out to additional regions in the near future

See OCPBUGS-11636 for additional information

Version-Release number of selected component (if applicable):

 

How reproducible:

Reproducible in affected regions.

Steps to Reproduce:

1. Use "ccoctl aws create-all" flow to create STS infrastructure in an affected region like us-east-2. Notice that document upload fails because the s3 bucket is created in a state that does not allow usage of ACLs with the s3 bucket.

Actual results:

./ccoctl aws create-all --name abutchertestue2 --region us-east-2 --credentials-requests-dir ./credrequests --output-dir _output
2023/04/11 13:01:06 Using existing RSA keypair found at _output/serviceaccount-signer.private
2023/04/11 13:01:06 Copying signing key for use by installer
2023/04/11 13:01:07 Bucket abutchertestue2-oidc created
2023/04/11 13:01:07 Failed to create Identity provider: failed to upload discovery document in the S3 bucket abutchertestue2-oidc: AccessControlListNotSupported: The bucket does not allow ACLs
        status code: 400, request id: 2TJKZC6C909WVRK7, host id: zQckCPmozx+1yEhAj+lnJwvDY9rG14FwGXDnzKIs8nQd4fO4xLWJW3p9ejhFpDw3c0FE2Ggy1Yc=

Expected results:

"ccoctl aws create-all" successfully creates IAM and S3 infrastructure. OIDC discovery and JWKS documents are successfully uploaded to the S3 bucket and are publicly accessible.

Additional info:

 

This is a clone of issue OCPBUGS-9464. The following is the description of the original issue:

Description of problem:

mtls connection is not working when using an intermetiate CA appart from the root CA, both with CRL defined.
The Intermediate CA Cert had a published CDP which directed to a CRL issued by the root CA.

The config map in the openshift-ingress namespace contains the CRL as issued by the root CA. The CRL issued by the Intermediate CA is not present since that CDP is in the user cert and so not in the bundle.

When attempting to connect using a user certificate issued by the Intermediate CA it fails with an error of unknown CA.

When attempting to connect using a user certificate issued by the to Root CA the connection is successful.

Version-Release number of selected component (if applicable):

4.10.24

How reproducible:
Always

Steps to Reproduce:

1. Configure CA and intermediate CA with CRL
2. Sign client certificate with the intermediate CA
3. Configure mtls in openshift-ingress

Actual results:

When attempting to connect using a user certificate issued by the Intermediate CA it fails with an error of unknown CA.
When attempting to connect using a user certificate issued by the to Root CA the connection is successful.

Expected results:

Be able to connect with client certificated signed by the intermediate CA

Additional info:

This bug is a backport clone of [Bugzilla Bug 2118318](https://bugzilla.redhat.com/show_bug.cgi?id=2118318). The following is the description of the original bug:

+++ This bug was initially created as a clone of Bug #2117569 +++

Description of problem:

The garbage collector resource quota controller must ignore ALL events; otherwise, if a rogue controller or a workload causes unbound event creation, performance will degrade as it has to process the events.

Fix: https://github.com/kubernetes/kubernetes/pull/110939

This bug is to track fix in master (4.12) and also allow to backport to 4.11.1

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:

Expected results:

Additional info:

— Additional comment from Michal Fojtik on 2022-08-11 10:52:28 UTC —

I'm using FastFix here as we need to backport this to 4.11.1 to avoid support churn for busy clusters or clusters doing upgrades.

— Additional comment from ART BZ Bot on 2022-08-11 15:13:32 UTC —

Elliott changed bug status from MODIFIED to ON_QA.
This bug is expected to ship in the next 4.12 release.

— Additional comment from zhou ying on 2022-08-12 03:03:34 UTC —

checked the payload commit id , the payload 4.12.0-0.nightly-2022-08-11-191750 has container the fixed pr .

oc adm release info registry.ci.openshift.org/ocp/release:4.12.0-0.nightly-2022-08-11-191750 --commit-urls |grep hyperkube
Warning: the default reading order of registry auth file will be changed from "${HOME}/.docker/config.json" to podman registry config locations in the future version. "${HOME}/.docker/config.json" is deprecated, but can still be used for storing credentials as a fallback. See https://github.com/containers/image/blob/main/docs/containers-auth.json.5.md for the order of podman registry config locations.
hyperkube https://github.com/openshift/kubernetes/commit/da80cd038ee5c3b45ba36d4b48b42eb8a74439a3

commit da80cd038ee5c3b45ba36d4b48b42eb8a74439a3 (HEAD -> master, origin/release-4.13, origin/release-4.12, origin/master, origin/HEAD)
Merge: a9d6306a701 055b96e614a
Author: OpenShift Merge Robot <openshift-merge-robot@users.noreply.github.com>
Date: Thu Aug 11 15:13:05 2022 +0000

Merge pull request #1338 from benluddy/openshift-pick-110888

Bug 2117569: UPSTREAM: 110888: feat: fix a bug thaat not all event be ignored by gc controller

Description of problem:

The ovn-kubernetes ovnkube-master containers are continuously crashlooping since we updated to 4.11.0-0.okd-2022-10-15-073651.

Log Excerpt:

] [] []  [{kubectl-client-side-apply Update networking.k8s.io/v1 2022-09-12 12:25:06 +0000 UTC FieldsV1 {"f:metadata":{"f:annotations":{".":{},"f:kubectl.kubernetes.io/last-applied-configuration":{}}},"f:spec":{"f:ingress":{},"f:policyTypes":{}}} }]},Spec:NetworkPolicySpec{PodSelector:{map[] []},Ingress:[]NetworkPolicyIngressRule{NetworkPolicyIngressRule{Ports:[]NetworkPolicyPort{},From:[]NetworkPolicyPeer{NetworkPolicyPeer{PodSelector:&v1.LabelSelector{MatchLabels:map[string]string{access: true,},MatchExpressions:[]LabelSelectorRequirement{},},NamespaceSelector:nil,IPBlock:nil,},},},},Egress:[]NetworkPolicyEgressRule{},PolicyTypes:[Ingress],},} &NetworkPolicy{ObjectMeta:{allow-from-openshift-ingress  compsci-gradcentral  a405f843-c250-40d7-8dd4-a759f764f091 217304038 1 2022-09-22 14:36:38 +0000 UTC <nil> <nil> map[] map[] [] []  [{openshift-apiserver Update networking.k8s.io/v1 2022-09-22 14:36:38 +0000 UTC FieldsV1 {"f:spec":{"f:ingress":{},"f:policyTypes":{}}} }]},Spec:NetworkPolicySpec{PodSelector:{map[] []},Ingress:[]NetworkPolicyIngressRule{NetworkPolicyIngressRule{Ports:[]NetworkPolicyPort{},From:[]NetworkPolicyPeer{NetworkPolicyPeer{PodSelector:nil,NamespaceSelector:&v1.LabelSelector{MatchLabels:map[string]string{policy-group.network.openshift.io/ingress: ,},MatchExpressions:[]LabelSelectorRequirement{},},IPBlock:nil,},},},},Egress:[]NetworkPolicyEgressRule{},PolicyTypes:[Ingress],},}]: cannot clean up egress default deny ACL name: error in transact with ops [{Op:mutate Table:Port_Group Row:map[] Rows:[] Columns:[] Mutations:[{Column:acls Mutator:delete Value:{GoSet:[{GoUUID:60cb946a-46e9-4623-9ba4-3cb35f018ed6}]}}] Timeout:<nil> Where:[where column _uuid == {ccdd01bf-3009-42fb-9672-e1df38190cd7}] Until: Durable:<nil> Comment:<nil> Lock:<nil> UUIDName:} {Op:mutate Table:Port_Group Row:map[] Rows:[] Columns:[] Mutations:[{Column:acls Mutator:delete Value:{GoSet:[{GoUUID:60cb946a-46e9-4623-9ba4-3cb35f018ed6}]}}] Timeout:<nil> Where:[where column _uuid == {10bbf229-8c1b-4c62-b36e-4ba0097722db}] Until: Durable:<nil> Comment:<nil> Lock:<nil> UUIDName:} {Op:delete Table:ACL Row:map[] Rows:[] Columns:[] Mutations:[] Timeout:<nil> Where:[where column _uuid == {7b55ba0c-150f-4a63-9601-cfde25f29408}] Until: Durable:<nil> Comment:<nil> Lock:<nil> UUIDName:} {Op:delete Table:ACL Row:map[] Rows:[] Columns:[] Mutations:[] Timeout:<nil> Where:[where column _uuid == {60cb946a-46e9-4623-9ba4-3cb35f018ed6}] Until: Durable:<nil> Comment:<nil> Lock:<nil> UUIDName:}] results [{Count:1 Error: Details: UUID:{GoUUID:} Rows:[]} {Count:1 Error: Details: UUID:{GoUUID:} Rows:[]} {Count:1 Error: Details: UUID:{GoUUID:} Rows:[]} {Count:1 Error: Details: UUID:{GoUUID:} Rows:[]} {Count:0 Error:referential integrity violation Details:cannot delete ACL row 7b55ba0c-150f-4a63-9601-cfde25f29408 because of 1 remaining reference(s) UUID:{GoUUID:} Rows:[]}] and errors []: referential integrity violation: cannot delete ACL row 7b55ba0c-150f-4a63-9601-cfde25f29408 because of 1 remaining reference(s)

Additional info:

https://github.com/okd-project/okd/issues/1372

Issue persisted through update to 4.11.0-0.okd-2022-10-28-153352

must-gather: https://nbc9-snips.cloud.duke.edu/snips/must-gather.local.2859117512952590880.zip

This is a clone of issue OCPBUGS-416. The following is the description of the original issue:

Description of problem:
The pod fails to mount the PVC using IBM Cloud VPC block storage.

Version-Release number of selected component (if applicable):

How reproducible:
The steps can be followed here: from this link
https://cloud.ibm.com/docs/openshift?topic=openshift-vpc-block
.
The error occurs when the application pod tried to mount the VPC.

Steps to Reproduce:
Describe above.

Actual results:
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 26m default-scheduler Successfully assigned default/test to a100-huge-m25p7-worker-3-with-secondary-xdwvl
Normal SuccessfulAttachVolume 26m attachdetach-controller AttachVolume.Attach succeeded for volume "pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba"
Warning FailedMount 26m (x2 over 26m) kubelet MountVolume.MountDevice failed for volume "pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba" : rpc error: code = Internal desc = {RequestID: ffbb97b4-e4d0-4016-87a9-dc46f80c5478 , Code: FormatAndMountFailed, Description: Failed to format '/dev/disk/by-id/virtio-0777-6872e22d-5c00-4' and mount it at '/var/lib/kubelet/plugins/kubernetes.io/csi/pv/pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba/globalmount', BackendError: format of disk "/dev/disk/by-id/virtio-0777-6872e22d-5c00-4" failed: type"ext4") target"/var/lib/kubelet/plugins/kubernetes.io/csi/pv/pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba/globalmount") options"defaults") errcode:(exit status 1) output:(mke2fs 1.45.6 (20-Mar-2020)
The file /dev/disk/by-id/virtio-0777-6872e22d-5c00-4 does not exist and no size was specified.
) , Action: Please check if there is any error in POD describe related with volume attach}
Warning FailedMount 22m kubelet Unable to attach or mount volumes: unmounted volumes=[bs-pvc], unattached volumes=[kube-api-access-6bgvj bs-pvc]: timed out waiting for the condition
Warning FailedMount 4m11s (x9 over 24m) kubelet Unable to attach or mount volumes: unmounted volumes=[bs-pvc], unattached volumes=[bs-pvc kube-api-access-6bgvj]: timed out waiting for the condition
Warning FailedMount 3m51s (x17 over 26m) kubelet MountVolume.MountDevice failed for volume "pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba" : rpc error: code = Internal desc = {RequestID: 1a12a7c5-3bd0-41cf-b8a9-90dd3224c2fb , Code: FormatAndMountFailed, Description: Failed to format '/dev/disk/by-id/virtio-0777-6872e22d-5c00-4' and mount it at '/var/lib/kubelet/plugins/kubernetes.io/csi/pv/pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba/globalmount', BackendError: format of disk "/dev/disk/by-id/virtio-0777-6872e22d-5c00-4" failed: type"ext4") target"/var/lib/kubelet/plugins/kubernetes.io/csi/pv/pvc-8721c341-739d-4607-bbcb-9dcf66ef6dba/globalmount") options"defaults") errcode:(exit status 1) output:(mke2fs 1.45.6 (20-Mar-2020)

Expected results:
The pod should successfully mount the PVC

Additional info:
Had a debugging session with Sameer Shaikh and Arashad Ahamad from the IBM VPC block storage team. The conclusion is that the udevadm utility is missing in the IPI image used by the IBM Cloud VPC block storage CSI.

  1. oc exec -it ibm-vpc-block-csi-controller-5fbb46bdc6-k7kpf -n openshift-cluster-csi-drivers -c iks-vpc-block-driver bash
    kubectl exec [POD] [COMMAND] is DEPRECATED and will be removed in a future version. Use kubectl exec [POD][COMMAND] instead.
    bash-4.4$ which udevadm
    which: no udevadm in (/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin)

Fixes rolled out for FIPS make it impossible for a binary compiled on RHELX to run on RHELY (in a FIPS compliant way) if the RHELY host is running with FIPS enabled.

This is because the binary will not find a compatible OpenSSL library and fall back to using internal Go crypto. Only OpenSSL is FIPS certified.

With the introduction of "FIPS or Die", binaries run in FIPS mode but which cannot use OpenSSL will exit with an error. Customers using RHEL7 with FIPS enabled, therefore, need a RHEL7 based oc.

This is a clone of issue OCPBUGS-7650. The following is the description of the original issue:

This is a clone of issue OCPBUGS-672. The following is the description of the original issue:

Description of problem:

Redhat-operator part of the marketplace is failing regularly due to startup probe timing out connecting to registry-server container part of the same pod within 1 sec which in turn increases CPU/Mem usage on Master nodes:

62m         Normal    Scheduled                pod/redhat-operators-zb4j7                         Successfully assigned openshift-marketplace/redhat-operators-zb4j7 to ip-10-0-163-212.us-west-2.compute.internal by ip-10-0-149-93
62m         Normal    AddedInterface           pod/redhat-operators-zb4j7                         Add eth0 [10.129.1.112/23] from ovn-kubernetes
62m         Normal    Pulling                  pod/redhat-operators-zb4j7                         Pulling image "registry.redhat.io/redhat/redhat-operator-index:v4.11"
62m         Normal    Pulled                   pod/redhat-operators-zb4j7                         Successfully pulled image "registry.redhat.io/redhat/redhat-operator-index:v4.11" in 498.834447ms
62m         Normal    Created                  pod/redhat-operators-zb4j7                         Created container registry-server
62m         Normal    Started                  pod/redhat-operators-zb4j7                         Started container registry-server
62m         Warning   Unhealthy                pod/redhat-operators-zb4j7                         Startup probe failed: timeout: failed to connect service ":50051" within 1s
62m         Normal    Killing                  pod/redhat-operators-zb4j7                         Stopping container registry-server


Increasing the threshold of the probe might fix the problem:
  livenessProbe:
      exec:
        command:
        - grpc_health_probe
        - -addr=:50051
      failureThreshold: 3
      initialDelaySeconds: 10
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 5
    name: registry-server
    ports:
    - containerPort: 50051
      name: grpc
      protocol: TCP
    readinessProbe:
      exec:
        command:
        - grpc_health_probe
        - -addr=:50051
      failureThreshold: 3
      initialDelaySeconds: 5
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 5 

Version-Release number of selected component (if applicable):

 

How reproducible:

Always

Steps to Reproduce:

1. Install OSD cluster using 4.11.0-0.nightly-2022-08-26-162248 payload
2. Inspect redhat-operator pod in openshift-marketplace namespace
3. Observe the resource usage ( CPU and Memory ) of the pod 

Actual results:

Redhat-operator failing leading to increase to CPU and Mem usage on master nodes regularly during the startup

Expected results:

Redhat-operator startup probe succeeding and no spikes in resource on master nodes

Additional info:

Attached cpu, memory and event traces.

 

Description of problem:

Intended to backport the corresponding https://bugzilla.redhat.com/show_bug.cgi?id=2095852 which has been fixed already for this version.

Version-Release number of selected component (if applicable):

 

How reproducible:

 

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-858. The following is the description of the original issue:

Description of problem:

In OCP 4.9, the package-server-manager was introduced to manage the packageserver CSV. However, when OCP 4.8 in upgraded to 4.9, the packageserver stays stuck in v0.17.0, which is the version in OCP 4.8, and v0.18.3 does not roll out, which is the version in OCP 4.9

Version-Release number of selected component (if applicable):

 

How reproducible:

Always

Steps to Reproduce:

1. Install OCP 4.8

2. Upgrade to OCP 4.9 

$ oc get clusterversion 
NAME      VERSION                             AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.8.0-0.nightly-2022-08-31-160214   True        True          50m     Working towards 4.9.47: 619 of 738 done (83% complete)

$ oc get clusterversion 
NAME      VERSION   AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.9.47    True        False         4m26s   Cluster version is 4.9.47
 

Actual results:

Check packageserver CSV. It's in v0.17.0 

$ oc get csv  NAME            DISPLAY          VERSION   REPLACES   PHASE packageserver   Package Server   0.17.0               Succeeded 

Expected results:

packageserver CSV is at 0.18.3 

Additional info:

packageserver CSV version in 4.8: https://github.com/openshift/operator-framework-olm/blob/release-4.8/manifests/0000_50_olm_15-packageserver.clusterserviceversion.yaml#L12

packageserver CSV version in 4.9: https://github.com/openshift/operator-framework-olm/blob/release-4.9/pkg/manifests/csv.yaml#L8

Description of problem:

Provisioning interface on master node not getting ipv4 dhcp ip address from bootstrap dhcp server on OCP 4.10.16 IPI BareMetal install.

Customer is performing an OCP 4.10.16 IPI BareMetal install and bootstrap node provisions just fine, but when master nodes are booted for provisioning, they are not getting an ipv4 address via dhcp. As such, the install is not moving forward at this point.

Version-Release number of selected component (if applicable):

OCP 4.10.16

How reproducible:

Perform OCP 4.10.16 IPI BareMetal install.

Actual results:

provisioning interface comes up (as evidenced by ipv6 address) but is not getting an ipv4 address via dhcp. OCP install / provisioning fails at this point.

Expected results:

provisioning interface successfully received an ipv4 ip address and successfully provisioned master nodes (and subsequently worker nodes as well.)

Additional info:

As a troubleshooting measure, manually adding an ipv4 ip address did allow the coreos image on the bootstrap node to be reached via curl.

Further, the kernel boot line for the first master node was updated for a static ip addresss assignment for further confirmation that the master node would successfully image this way which further confirming that the issue is the provisioning interface not receiving an ipv4 ip address from the dhcp server.

Description of problem:

Requesting backport for the pull request  https://github.com/openshift/console/pull/12109 in RHOCP 4.11

Version-Release number of selected component (if applicable):

4.11.z

How reproducible:

NA

Steps to Reproduce:

1.
2.
3.

Actual results:

Developer view does not display Logs tab

Expected results:

Developer view should display Logs tab

Additional info:

Jira to be referred https://issues.redhat.com/browse/LOG-3388

Description of problem:

If you set a services cluster IP to an IP with a leading zero (e.g. 192.168.0.011), ovn-k should normalise this and remove the leading zero before sending it to ovn.

This was seen by me on a CI run executing the k8 test here: test/e2e/network/funny_ips.go +75

you can reproduce using that above test.

Have a read of the text there:

 43 // What are funny IPs:  
 44 // The adjective is because of the curl blog that explains the history and the problem of liberal  
 45 // parsing of IP addresses and the consequences and security risks caused the lack of normalization,
 46 // mainly due to the use of different notations to abuse parsers misalignment to bypass filters.
 47 // xref: https://daniel.haxx.se/blog/2021/04/19/curl-those-funny-ipv4-addresses/   
 48 //     
 49 // Since golang 1.17, IPv4 addresses with leading zeros are rejected by the standard library.
 50 // xref: https://github.com/golang/go/issues/30999
 51 //     
 52 // Because this change on the parsers can cause that previous valid data become invalid, Kubernetes
 53 // forked the old parsers allowing leading zeros on IPv4 address to not break the compatibility.
 54 //     
 55 // Kubernetes interprets leading zeros on IPv4 addresses as decimal, users must not rely on parser
 56 // alignment to not being impacted by the associated security advisory: CVE-2021-29923 golang
 57 // standard library "net" - Improper Input Validation of octal literals in golang 1.16.2 and below
 58 // standard library "net" results in indeterminate SSRF & RFI vulnerabilities. xref:
 59 // https://nvd.nist.gov/vuln/detail/CVE-2021-29923                                                                                                     

northd is logging an error about this also:

|socket_util|ERR|172.30.0.011:7180: bad IP address "172.30.0.011" 
...
2022-08-23T14:14:21.968Z|01839|ovn_util|WARN|bad ip address or port for load balancer key 172.30.0.011:7180

 

Also, I see the error:

E0823 14:14:34.135115    3284 gateway_shared_intf.go:600] Failed to delete conntrack entry for service e2e-funny-ips-8626/funny-ip: failed to delete conntrack entry for service e2e-funny-ips-8626/funny-ip with svcVIP 172.30.0.011, svcPort 7180, protocol TCP: value "<nil>" passed to DeleteConntrack is not an IP address 

We should normalise the IPs before sending to OVN-k. I see also theres conntrack error when trying to set this bad IP.

 

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. See above k8 test

Actual results:

Leading zero IP sent to OVN

Expected results:

No leading zero IP sent to OVN

Additional info:

terraform-provider-libvirt uses SPICE by default for its libvirtxml.DomainGraphic, but this has been deprecated/removed in RHEL9/CentOS9 Stream.

Check https://github.com/openshift/installer/pull/6062

This is a clone of issue OCPBUGS-5019. The following is the description of the original issue:

This is a clone of issue OCPBUGS-4941. The following is the description of the original issue:

Description of problem: This is a follow-up to OCPBUGS-3933.

The installer fails to destroy the cluster when the OpenStack object storage omits 'content-type' from responses, and a container is empty.

Version-Release number of selected component (if applicable):

4.8.z

How reproducible:

Likely not happening in customer environments where Swift is exposed directly. We're seeing the issue in our CI where we're using a non-RHOSP managed cloud.

Steps to Reproduce:

1.
2.
3.

Actual results:

 

Expected results:

 

Additional info:

 

This is a clone of issue OCPBUGS-676. The following is the description of the original issue:

the machine approver isn't recognizing hostnames that use capital letters as valid even though DNS is case-insensitive

an example of this is in OHSS-14709:

I0822 19:04:51.587266       1 controller.go:114] Reconciling CSR: csr-vdtpv
I0822 19:04:51.600941       1 csr_check.go:156] csr-vdtpv: CSR does not appear to be client csr
I0822 19:04:51.603648       1 csr_check.go:542] retrieving serving cert from ip-100-66-119-117.ec2.internal (100.66.119.117:10250)
I0822 19:04:51.604003       1 csr_check.go:181] Failed to retrieve current serving cert: dial tcp 100.66.119.117:10250: connect: connection refused
I0822 19:04:51.604017       1 csr_check.go:201] Falling back to machine-api authorization for ip-100-66-119-117.ec2.internal
E0822 19:04:51.604024       1 csr_check.go:392] csr-vdtpv: DNS name 'ip-100-66-119-117.tech-ace-maint-prd.aws.delta.com' not in machine names: ip-100-66-119-117.ec2.internal ip-100-66-119-117.ec2.internal ip-100-66-119-117.tech-ACE-maint-prd.aws.delta.com
I0822 19:04:51.604033       1 csr_check.go:204] Could not use Machine for serving cert authorization: DNS name 'ip-100-66-119-117.tech-ace-maint-prd.aws.delta.com' not in machine names: ip-100-66-119-117.ec2.internal ip-100-66-119-117.ec2.internal ip-100-66-119-117.tech-ACE-maint-prd.aws.delta.com
I0822 19:04:51.606777       1 controller.go:199] csr-vdtpv: CSR not authorized

This can be worked around by manually approving the CSR

The relevant line in the machine approver appears to be here: https://github.com/openshift/cluster-machine-approver/blob/master/pkg/controller/csr_check.go#L378

Please review the following PR: https://github.com/openshift/csi-driver-nfs/pull/129

The PR has been automatically opened by ART (#forum-ocp-art) team automation and indicates
that the image(s) being used downstream for production builds are not consistent
with the images referenced in this component's github repository.

Differences in upstream and downstream builds impact the fidelity of your CI signal.

If you disagree with the content of this PR, please contact @release-artists
in #forum-ocp-art to discuss the discrepancy.

Closing this issue without addressing the difference will cause the issue to
be reopened automatically.

Tracker issue for bootimage bump in 4.11. This issue should block issues which need a bootimage bump to fix.

The previous bump was OCPBUGS-562.

Description of problem:

Create Loadbalancer type service within the OCP 4.11.x OVNKubernetes cluster to expose the api server endpoint, the service does not response for normal oc request. 
But some of them are working, like "oc whoami", "oc get --raw /api"

Version-Release number of selected component (if applicable):

4.11.8 with OVNKubernetes

How reproducible:

always

Steps to Reproduce:

1. Setup openshift cluster 4.11 on AWS with OVNKubernetes as the default network
2. Create the following service under openshift-kube-apiserver namespace to expose the api
----
apiVersion: v1
kind: Service
metadata:
  annotations:
    service.beta.kubernetes.io/aws-load-balancer-connection-idle-timeout: "1800"
  finalizers:
  - service.kubernetes.io/load-balancer-cleanup
  name: test-api
  namespace: openshift-kube-apiserver
spec:
  allocateLoadBalancerNodePorts: true
  externalTrafficPolicy: Cluster
  internalTrafficPolicy: Cluster
  ipFamilies:
  - IPv4
  ipFamilyPolicy: SingleStack
  loadBalancerSourceRanges:
  - <my_ip>/32
  ports:
  - nodePort: 31248
    port: 6443
    protocol: TCP
    targetPort: 6443
  selector:
    apiserver: "true"
    app: openshift-kube-apiserver
  sessionAffinity: None
  type: LoadBalancer

3. Setup the DNS resolution for the access
xxx.mydomain.com ---> <elb-auto-generated-dns>

4. Try to access the cluster api via the service above by updating the kubeconfig to use the custom dns name

Actual results:

No response from the server side.

$ time oc get node -v8
I1025 08:29:10.284069  103974 loader.go:375] Config loaded from file:  bmeng.kubeconfig
I1025 08:29:10.294017  103974 round_trippers.go:420] GET https://rh-api.bmeng-ccs-ovn.3o13.s1.devshift.org:6443/api/v1/nodes?limit=500
I1025 08:29:10.294035  103974 round_trippers.go:427] Request Headers:
I1025 08:29:10.294043  103974 round_trippers.go:431]     Accept: application/json;as=Table;v=v1;g=meta.k8s.io,application/json;as=Table;v=v1beta1;g=meta.k8s.io,application/json
I1025 08:29:10.294052  103974 round_trippers.go:431]     User-Agent: oc/openshift (linux/amd64) kubernetes/e40bd2d
I1025 08:29:10.365119  103974 round_trippers.go:446] Response Status: 200 OK in 71 milliseconds
I1025 08:29:10.365142  103974 round_trippers.go:449] Response Headers:
I1025 08:29:10.365148  103974 round_trippers.go:452]     Audit-Id: 83b9d8ae-05a4-4036-bff6-de371d5bec12
I1025 08:29:10.365155  103974 round_trippers.go:452]     Cache-Control: no-cache, private
I1025 08:29:10.365161  103974 round_trippers.go:452]     Content-Type: application/json
I1025 08:29:10.365167  103974 round_trippers.go:452]     X-Kubernetes-Pf-Flowschema-Uid: 2abc2e2d-ada3-4cb8-a86f-235df3a4e214
I1025 08:29:10.365173  103974 round_trippers.go:452]     X-Kubernetes-Pf-Prioritylevel-Uid: 02f7a188-43c7-4827-af58-5ebe861a1891
I1025 08:29:10.365179  103974 round_trippers.go:452]     Date: Tue, 25 Oct 2022 08:29:10 GMT
^C
real    17m4.840s
user    0m0.567s
sys    0m0.163s


However, it has the correct response if using --raw to request, eg:
$ oc get --raw /api/v1  --kubeconfig bmeng.kubeconfig 
{"kind":"APIResourceList","groupVersion":"v1","resources":[{"name":"bindings","singularName":"","namespaced":true,"kind":"Binding","verbs":["create"]},{"name":"componentstatuses","singularName":"","namespaced":false,"kind":"ComponentStatus","verbs":["get","list"],"shortNames":["cs"]},{"name":"configmaps","singularName":"","namespaced":true,"kind":"ConfigMap","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["cm"],"storageVersionHash":"qFsyl6wFWjQ="},{"name":"endpoints","singularName":"","namespaced":true,"kind":"Endpoints","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["ep"],"storageVersionHash":"fWeeMqaN/OA="},{"name":"events","singularName":"","namespaced":true,"kind":"Event","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["ev"],"storageVersionHash":"r2yiGXH7wu8="},{"name":"limitranges","singularName":"","namespaced":true,"kind":"LimitRange","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["limits"],"storageVersionHash":"EBKMFVe6cwo="},{"name":"namespaces","singularName":"","namespaced":false,"kind":"Namespace","verbs":["create","delete","get","list","patch","update","watch"],"shortNames":["ns"],"storageVersionHash":"Q3oi5N2YM8M="},{"name":"namespaces/finalize","singularName":"","namespaced":false,"kind":"Namespace","verbs":["update"]},{"name":"namespaces/status","singularName":"","namespaced":false,"kind":"Namespace","verbs":["get","patch","update"]},{"name":"nodes","singularName":"","namespaced":false,"kind":"Node","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["no"],"storageVersionHash":"XwShjMxG9Fs="},{"name":"nodes/proxy","singularName":"","namespaced":false,"kind":"NodeProxyOptions","verbs":["create","delete","get","patch","update"]},{"name":"nodes/status","singularName":"","namespaced":false,"kind":"Node","verbs":["get","patch","update"]},{"name":"persistentvolumeclaims","singularName":"","namespaced":true,"kind":"PersistentVolumeClaim","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["pvc"],"storageVersionHash":"QWTyNDq0dC4="},{"name":"persistentvolumeclaims/status","singularName":"","namespaced":true,"kind":"PersistentVolumeClaim","verbs":["get","patch","update"]},{"name":"persistentvolumes","singularName":"","namespaced":false,"kind":"PersistentVolume","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["pv"],"storageVersionHash":"HN/zwEC+JgM="},{"name":"persistentvolumes/status","singularName":"","namespaced":false,"kind":"PersistentVolume","verbs":["get","patch","update"]},{"name":"pods","singularName":"","namespaced":true,"kind":"Pod","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["po"],"categories":["all"],"storageVersionHash":"xPOwRZ+Yhw8="},{"name":"pods/attach","singularName":"","namespaced":true,"kind":"PodAttachOptions","verbs":["create","get"]},{"name":"pods/binding","singularName":"","namespaced":true,"kind":"Binding","verbs":["create"]},{"name":"pods/ephemeralcontainers","singularName":"","namespaced":true,"kind":"Pod","verbs":["get","patch","update"]},{"name":"pods/eviction","singularName":"","namespaced":true,"group":"policy","version":"v1","kind":"Eviction","verbs":["create"]},{"name":"pods/exec","singularName":"","namespaced":true,"kind":"PodExecOptions","verbs":["create","get"]},{"name":"pods/log","singularName":"","namespaced":true,"kind":"Pod","verbs":["get"]},{"name":"pods/portforward","singularName":"","namespaced":true,"kind":"PodPortForwardOptions","verbs":["create","get"]},{"name":"pods/proxy","singularName":"","namespaced":true,"kind":"PodProxyOptions","verbs":["create","delete","get","patch","update"]},{"name":"pods/status","singularName":"","namespaced":true,"kind":"Pod","verbs":["get","patch","update"]},{"name":"podtemplates","singularName":"","namespaced":true,"kind":"PodTemplate","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"storageVersionHash":"LIXB2x4IFpk="},{"name":"replicationcontrollers","singularName":"","namespaced":true,"kind":"ReplicationController","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["rc"],"categories":["all"],"storageVersionHash":"Jond2If31h0="},{"name":"replicationcontrollers/scale","singularName":"","namespaced":true,"group":"autoscaling","version":"v1","kind":"Scale","verbs":["get","patch","update"]},{"name":"replicationcontrollers/status","singularName":"","namespaced":true,"kind":"ReplicationController","verbs":["get","patch","update"]},{"name":"resourcequotas","singularName":"","namespaced":true,"kind":"ResourceQuota","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["quota"],"storageVersionHash":"8uhSgffRX6w="},{"name":"resourcequotas/status","singularName":"","namespaced":true,"kind":"ResourceQuota","verbs":["get","patch","update"]},{"name":"secrets","singularName":"","namespaced":true,"kind":"Secret","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"storageVersionHash":"S6u1pOWzb84="},{"name":"serviceaccounts","singularName":"","namespaced":true,"kind":"ServiceAccount","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["sa"],"storageVersionHash":"pbx9ZvyFpBE="},{"name":"serviceaccounts/token","singularName":"","namespaced":true,"group":"authentication.k8s.io","version":"v1","kind":"TokenRequest","verbs":["create"]},{"name":"services","singularName":"","namespaced":true,"kind":"Service","verbs":["create","delete","deletecollection","get","list","patch","update","watch"],"shortNames":["svc"],"categories":["all"],"storageVersionHash":"0/CO1lhkEBI="},{"name":"services/proxy","singularName":"","namespaced":true,"kind":"ServiceProxyOptions","verbs":["create","delete","get","patch","update"]},{"name":"services/status","singularName":"","namespaced":true,"kind":"Service","verbs":["get","patch","update"]}]}
 

Expected results:

The normal oc request should be working.

Additional info:

There is no such issue for clusters with openshift-sdn with the same OpenShift version and same LoadBalancer service.

We suspected that it might be related to the MTU setting, but this cannot explain why OpenShiftSDN works well.

Another thing might be related is that the OpenShiftSDN is using iptables for service loadbalancing and OVN is dealing that within the OVN services.

 

Please let me know if any debug log/info is needed.

Description of problem:
This is a clone of https://issues.redhat.com/browse/OCPBUGS-469

 

Description of problem: Numerous erroreneous logs in OVN master

I0823 18:00:11.163491       1 obj_retry.go:1063] Retry object setup: *v1.Pod openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k
I0823 18:00:11.163546       1 obj_retry.go:1096] Removing old object: *v1.Pod openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k
I0823 18:00:11.163555       1 pods.go:124] Deleting pod: openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k
I0823 18:00:11.163631       1 obj_retry.go:1103] Retry delete failed for *v1.Pod openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k, will try again later: deleteLogicalPort failed for pod openshift-operator-lifecycle-manager_collect-profiles-27687900-hlp6k: unable to locate portUUID+nodeName for pod openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k: error getting logical port <nil>: object not found
W0823 18:00:41.163633       1 obj_retry.go:1031] Dropping retry entry for *v1.Pod openshift-operator-lifecycle-manager/collect-profiles-27687900-hlp6k: exceeded number of failed attempts

Must-gather: http://shell.lab.bos.redhat.com/~anusaxen/must-gather.local.2234927131259452300/

 

Version-Release number of selected component (if applicable): 4.12.0-0.nightly-2022-08-23-031342

How reproducible: Always

Steps to Reproduce:
1. Bring up OVN cluster on 4.12
2.
3.

Actual results: deleteLogicalPort failed for already gone object

Expected results: deleteLogicalPort should not keep retrying post object deletion

Additional info:

This is a clone of issue RHIBMCS-151. The following is the description of the original issue:

Error msg

type: 'Warning' reason: 'ResolutionFailed' constraints not satisfiable: @existing/ibm-common-services//ibm-namespace-scope-operator.v2.0.0 and @existing/ibm-common-services//ibm-namespace-scope-operator.v1.15.0 provide NamespaceScope (operator.ibm.com/v1), subscription ibm-namespace-scope-operator requires @existing/ibm-common-services//ibm-namespace-scope-operator.v2.0.0, subscription ibm-namespace-scope-operator exists, clusterserviceversion ibm-namespace-scope-operator.v1.15.0 exists and is not referenced by a subscription

 

The issue happens  during the upgrade with and without channel switch. And several places which reports this issue

https://ibm-cloudplatform.slack.com/archi