Back to index
4.9.0-0.ci-2023-02-22-182045
Download the installer for your operating system or run
oc adm release extract --tools registry.ci.openshift.org/ocp/release:4.9.0-0.ci-2023-02-22-182045
Tests:
- Blocking jobs
- Informing jobs
Upgrades from:
Upgrades to:
Loading changelog, this may take a while ...
Created: 2023-02-22 18:28:47 +0000 UTC
Image Digest: sha256:3a2f336513080bcd350c3b1f41ace0b2710e81e49a327af73519174fd73c6178
Components
New images
Rebuilt images without code change
- bootstrap-kube-scheduler: specify resources.requests #445
- library-go vendoring for installer cmd timeout #444
- Bump openshift/api to bring the MatchLabelKeysInPodTopologySpread enabled in TPNoUpgrade #442
- trt-589 bump library-go #441
- Bump go.mod golang version to 1.19 #440
- k8s 1.25.0 #438
- Describe how to increase log level of kube-scheduler instances #436
- Updating ose-cluster-kube-scheduler-operator images to be consistent with ART #433
- Update owners #432
- Bug 2062459: Introduce sync unit test #430
- Bug 2086092: bump to k8s v0.24.0 #420
- Bug 2062459: Fix bootstrap leader election config #428
- Bug 2064024: README: fix scheduler configuration formatting #427
- Bug 2064024: Update library-go to 80f9619c2 #425
- Fix debugging information #424
- manifests/deployment: comply to restricted pod security level #421
- [rebase v1.24] remove –port flag from bootstrap #422
- Update OWNERS #419
- Updating ose-cluster-kube-scheduler-operator images to be consistent with ART #406
- Bug 2053582: Track static pod lifecycle #417
- Bug 2053582: Track static pod lifecycle #415
- Update resourcelock to configmapleases #412
- Bump(openshift/api): to get CSI changes #411
- Do not wait for a port which is no longer used by the scheduler #410
- update cert injection annotations to beta #409
- Bug 2029470: update library-go to get rapid installer pod fixes #407
- Bug 2035986: Replace deprecated annotation #393
- Bug 2042049: Delete feature gates removed from k8s.io/kubernetes/pkg/features/kube_features.go #401
- Bug 2039414: Update KubeSchedulerConfig to v1beta3(default in 1.23) #395
- Bug 2005901: Sync the library-go #397
- Bug 2033751: Return Error when trying to use Scheduler Policy #390
- Bug 2040132: drop –port flag #398
- Bug 2035927: Replace NodeResourcesLeastAllocated/NodeResourcesMostAllocated plugins with NodeResourcesFit #392
- Bug 2005901: Bump library-go #396
- Bug 2038968: bump(openshift/api): to get CSI migration gates switched off #394
- Bug 2005901: Guard controller pdb #373
- Bug 2034515: bump library-go #388
- Upgrade library-go version #387
- Bump to k8s 1.23 #386
- bump(library-go): to gather a fix for static pod UID version #384
- Bug 2026352: Sync with library-go to pick fixes for pruner panic #383
- Bug 2002300: Disable balancedAllocation and add weight for HighNodeUtilization profile #378
- update scheduler metrics service selector #382
- Fix gomod #377
- Updating ose-cluster-kube-scheduler-operator images to be consistent with ART #375
- bump(*): library-go #376
- bump(openshift/api): to get fix for typo in feature gate #374
- Bug 2010354: Make alert consistent with the alerting-consistency guidelines #372
- podsecurity: enforce privileged for openshift-kube-scheduler namespace #371
- Updating ose-cluster-kube-scheduler-operator images to be consistent with ART #369
- Bug 2001856: bump libgo and related deps #368
- Updating ose-cluster-kube-scheduler-operator images to be consistent with ART #367
- Full changelog
- Updating ose-cluster-samples-operator images to be consistent with ART #465
- Add client certificate and key to service monitor #464
- Updating ose-cluster-samples-operator images to be consistent with ART #435
- Bug 2086086: Update Cluster Sample Operator dependencies and libraries for OCP 4.11 #433
- update jenkins CPaaS image refs prior to 4.11 GA #432
- Bug 2086086: Update Cluster Sample Operator dependencies and libraries for OCP 4.11 #431
- Bug 2095256: Samples Owner needs to be Updated #429
- Bug 2086086: Update Cluster Sample Operator dependencies and libraries for OCP 4.11 #428
- AUTH-133: manifests/deployment: comply to restricted pod security level #425
- Updating ose-cluster-samples-operator images to be consistent with ART #426
- JNKS-289: pull in jenkins imagestream updates (add back maven/nodejs streams) #422
- Jira SO-19: Make sure template and imagestream api version is groupified #420
- JNKS-287: remove imagestream manifest refs; remove override of jenkins images with payload images #416
- Bug 2010364: OpenShift Alerting Rules Style-Guide Compliance #419
- Bug 2067823: Taking care of CVE-2022-21698 #418
- Bug 2064610: Remove duplicate v1 from cakephp-mysql templates #417
- manifests: Add capability.openshift.io/name #414
- Updating ose-cluster-samples-operator images to be consistent with ART #412
- Bug 2027745: Allowing ImageStream creation when config registry empty #406
- Bug 2033720: Synching SSO library #411
- upd rhel7 Dockerfile in case still used for rhel7 worker nodes #410
- Bug 2033720: Bump k8 dependencies to v0.23.1 #409
- Bug 2033720: Library synchronization for OCP 4.10 #408
- Bug 2027745: Move openshift build-machinery-go dependency #407
- Update OWNERS #402
- Remove single node prod cluster profile annotation as it is not used #405
- Bug 2007757: Including Template in must-gather related objects. #399
- Bug 2006947: fix proxy portion of tbr inaccessible check #397
- Updating ose-cluster-samples-operator images to be consistent with ART #396
- Bug 2002368: acccount for image api returning invalid on imagestream create based on allowed/blocked registry settings #394
- Bug 2003683: fix recently introduced panic when conflict errors occur #395
- Bug 1993840: more avoiding immediate degraded on intermittent API errors #391
- Updating ose-cluster-samples-operator images to be consistent with ART #390
- Full changelog
- Clean up owners file as part of DPP-10343 #46
- Updating ose-cluster-update-keys images to be consistent with ART #45
- Updating ose-cluster-update-keys images to be consistent with ART #43
- Update OWNERS #44
- OWNERS: Prune crawford #42
- Full changelog
- Bump openshift/k8s-prometheus-adapter to v0.10.0 #64
- cmd/adapter: recover klog flags #524
- Cut release 0.10.0 #523
- Add –prometheus-verb to support POST requests to prometheus servers #509
- Update dependencies #521
- Update README.md with Helm-3 Command #491
- Change apiregistration.k8s.io to v1 #494
- fix: close file #498
- Validate OWNERS files #482
- Drop unused alias in OWNERS_ALIASES #476
- images: use k8s-staging-test-infra/gcb-docker-gcloud #472
- *: merge changes from v0.9.1 #461
- Add s to metricQuery #465
- Fix external metrics provider not respecting metrics-max-age #455
- docs: follow kubernetes/k8s.io branch rename: #454
- Add bug template #438
- README: improve gcr.k8s.io instructions #447
- v0.9.0: Bump version file to reflect new release #448
- OWNERS: Add Joao and myself, and move former team members to emeritus #63
- Updating ose-prometheus-adapter images to be consistent with ART #62
- Updating ose-prometheus-adapter images to be consistent with ART #61
- Updating ose-prometheus-adapter images to be consistent with ART #60
- Bump openshift/k8s-prometheus-adapter to v0.9.1 #59
- *: cut release-0.9.1 #460
- Fix external metrics provider not respecting metrics-max-age #459
- Updating ose-prometheus-adapter images to be consistent with ART #58
- OWNERS: cleanup #57
- Updating ose-prometheus-adapter images to be consistent with ART #56
- Full changelog
- OCPBUGS-69: Generate egress ip metrics for sdn node #470
- OCPBUGS-1533: rebase to kube 1.25.1 #458
- OCPBUGS-256: fix network policy egress #459
- Bug 2104953: rebase to 1.24, take 2 #450
- Enable EgressIP on OpenStack #447
- Bug 2081562: setup network policy rules during pod creation to fix postStart hook #439
- Updating kube-proxy images to be consistent with ART #448
- Bug 2101622: Revert ‘kube 1.24 rebase’; master #442
- Bug 2079958: Add TopologyAwareHints support in kube-proxy. #426
- Bug 2089350: kube 1.24 rebase #440
- Bug 2067865: Bump prometheus/client_golang to 1.11.1 #432
- Updating kube-proxy images to be consistent with ART #429
- Updating ose-sdn images to be consistent with ART #427
- Bug 2052332: use pods cache during reattach processing #411
- Bug 2016534: Exclude the default drop bit from egress IP VNID #428
- Bug 2016534: Masquerade in cluster traffic that is marked for egress IP #423
- Bug 2063123: Remove node-tainting for too-small MTU #417
- Bug 2050230: Use pagination when listing Netnamespaces #419
- Bug 2062558: egressip: Continue to process other nodes if a node is not ready #418
- Add unit tests for vxlan no track ip table rule #401
- Bug 2061919: Fix releasing egress IP in cloud environments #409
- Bug 2061916: mixed ingress and egress policies can result in half-isolated pods #408
- Bug 2060553: another fix for mixed ingress and egress policies #406
- Bug 2050230: Use pagination when listing resources #400
- Bug 2060553: Separate the allPodsSelected into egress and ingress #404
- Rebase SDN k8 1.23.4 #399
- clean up OWNERS #392
- Updating ose-sdn images to be consistent with ART #396
- Updating kube-proxy images to be consistent with ART #397
- move to-be-deleted userspace proxy code into sdn tree #393
- Bug 2043802: Fix CloudPrivateIPConfig enqueueing during egressip reassignment #395
- Bug 2038793: Use the kubeClient instead of the informer cache to fetch nodes for egress IP validation. #389
- Bug 2039294: SDN controller metrics cannot be scraped by prometheus #388
- Bug 2035439: Use cloud egress network config for verifying egress IP in cloud environment #387
- Bug 2036622: Fix CloudPrivateIPConfig sync on sdn-controller restart #386
- Handle MTU migration #381
- Fix flaky check for capacity test for egress IP, and better doc blocs #383
- Egress IP on public clouds #365
- Block Alibaba metadata service IP #380
- Adding
--platform-type argument #379
- NetworkPolicy ipblock except, port ranges #374
- adding network policy egress to openshift-sdn #372
- Bug 2024880: [EgressIP] move
ct(commit) action from OVS group to flow #373
- NP updates preparing for egress support #350
- update for ART #369
- WIP: Metrics: Add server for Egress IP/firewall #358
- Run clearInitialNodeNetworkUnavailableCondition earlier #364
- Add documentation about proxying/unidling #361
- Bug 2008987: Remove locking from EgressIPTracker.Ping #359
- Replace utilruntime.HandleError() with klog.Errorf() #356
- Updating kube-proxy images to be consistent with ART #351
- Bug 2003859: Fix up event recorder usage (again) #349
- resync proxy startup code with upstream #352
- remove a bunch of dead/useless code #345
- Updating ose-sdn images to be consistent with ART #344
- Bug 1966521: proxy: don’t re-check every userspace proxy rule on every change #342
- Fix Dockerfile.fedora #333
- README minor improvements #336
- Full changelog
- Merg upstream into downstream #60
- go, kind, dep version bump #178
- Tests e2e reduce abstraction #176
- Minor Pre-Acceptance: Set default method, return 500s on broken path regex. #174
- README.md: Emphasize pattern matching #173
- check all headers, go version bump #171
- github/workflows: fix golang setup for publish job #167
- *: cut v0.12.0 release #166
- add update-go-deps and perform initial update to k8s v0.23.1 #160
- Revert “Add Makefile target for go dep updates” #159
- Add RELEASE.md #156
- Add Makefile target for go dep updates #155
- Fix High/Medium CVEs (Common Vulnerabilities and Exposures) #149
- fix: use supported apiVersion for deployment and authorization #150
- Fix ppc64le image build which in turn fixed multi-arch build #147
- build.yml: increase linter timeout #144
- chore: move –upstream-force-h2c out of –insecure-listen-address #139 #140
- Bump dependency to support building on riscv64 arch #141
- OWNERS: add myself (ibihim) #57
- Updating kube-rbac-proxy images to be consistent with ART #54
- Updating kube-rbac-proxy images to be consistent with ART #52
- Full changelog
- Show warnings if provider specs have invalid GroupVersionKind #1018
- Update envtest #1071
- OCPCLOUD-1131: Add DescribeInstanceTypes permission for AWS provider #1062
- OCPBUGS-2151: Don’t degrade when workers not expected #1074
- Updating ose-machine-api-operator images to be consistent with ART #1073
- Update default Azure image to use image galleries #1076
- OCPBUGS-1274: add tolerations to termination handler #1072
- Restrict creating Power VS machine with negative processor and memory #1067
- OCPBUGS-519: update error message for disconnected installation validation #1068
- Update dependencies to K8s 1.25 #1070
- Only drain a single control plane machine at once #1066
- Add tag and category id to the caching client log messages #1063
- Bug 2101736: Ignore managed fields during finalizer removal #1061
- Ensure the phase transition metric only updates when the phase changes #1059
- Release leader election lease when shutting down #1060
- Tags/categories ids cache implementation #1057
- OCPBUGS-268: vSphere - enable steal time accounting #1056
- Bug 2111972: operator NS manifest: Set empty openshift.io/run-level #1053
- Change the git command #1050
- add a helper function to detect windows os #1055
- Bug 2115308: Ensure failed drains are subject to exponential backoff #1051
- Bug 2087981: Change “create” sequence with powering on the vm after clone #1047
- update windows helper functions #1049
- Bug 2110501: Revert “Bug 2101880: operator NS manifest: Set empty openshift.io/run-level” #1045
- Bug 2109258: add more support for old delete annotation #1042
- Bug 2101736: allow to remove finalizers #1037
- Bug 2107578: Fix while setting default processor value for Power VS platform #1040
- Bug 2104642: fix the Machine validation webhook for nutanix providerSpec #1038
- Bug 2104642: Add a validation webhook for Nutanix MachineProviderConfig #1034
- Bug 2097153: change ListTags call to ListTagsForCategory #1027
- Bug 1994820: Degrade operator on cluster bootstrap if not all Machines are Running #1019
- Migrate delete annotation to match upstream CAPI annotation #1024
- add utility functions for windows powershell tags #1030
- Bug 2101880: operator NS manifest: Set empty openshift.io/run-level #1031
- Add webhook validations for Azure boot diagnostics #1029
- Update revendoring tool #1028
- Bug 2082667: Separate controller for the node draining #1023
- Bug 2094196: Added webhook support for PowerVS platform #998
- Bug 2087039: Bump go to 1.18 #1022
- Bug 2087039: Update dependencies to K8s 1.24 #1021
- Updating ose-machine-api-operator images to be consistent with ART #1015
- Bug 2059338: Add template HW version detection during clone #1016
- Bug 2083237: Set vCenter client request timeout #1013
- Skip validation if we just remove the finalizer #996
- Don’t force a requeue after updating Status.Phase #1012
- Remove unreleased AWS Placement Groups support #1009
- Fix broken link to CVO run levels #991
- AWS IMDSv2 machine spec validation #1008
- Add DescribeRegions permission for aws controller #1007
- Add Nutanix Platform to Machine API Operator #988
- Add pd-balanced disks support for GCP #1006
- Adding cluster api provider url(ibm) #1005
- Refactor provider status to use metav1.Condition #1004
- webhook: machine: add Azure Data Disks deletionPolicy validation #1003
- Remove 30s wait before creating a machine #999
- Make MAPO the default for OpenStack deployments #1000
- Bug 2067852: Bump prometheus/client_golang #1002
- Machine webhook validation for Azure Ultra Disks #1001
- Update govmomi to 0.27.4 #995
- Bug 2060697: Update openshift/api dependency #994
- Fix MAPO image #993
- Update OpenShift API to include placement groups #992
- Replace broken image references #881
- modify prometheus alerts to ignore pod #986
- Updating ose-machine-api-operator images to be consistent with ART #983
- Allow to remove ec2 placement groups #990
- Validate AWS NetworkInterfaceType in webhook #989
- Add IAM permissions required for AWS Placement Group enhancement #976
- fix readme after migrating from cluster-api-provider-* #982
- Bug 2041971: Bump library-go to fix mutating webhook reconcile #981
- Bug 2042366: Update openshift/api to include list-map fixes for lifecycle hooks #979
- Bug 2040620: Rename MAPO image #978
- Bug 1918005: Use known vSphere cluster to uniquely identify networks. #961
- Bug 2033390: connect real MAPO images #975
- Bug 2001008: Change default cloneMode to fullClone. #959
- Bug 2034563: validate azure ephemeral storage location and caching type #974
- Bug 2034879: Enforce unique names for lifecycle hooks via webhooks #973
- Bug 2034484: feat: updated library-go #972
- Bug 2032926: Update dependencies to K8s 1.23 #971
- Bug 2011668: Improve powerOff task handling during machine deletion on vsphere platform #960
- AlibabaCloud: adding alibaba cloud platform #926
- Bug 2032296: bump openshift/api #969
- Bug 2031858: stop using beta pre-defined role for compute.targetPools #950
- Manifest syncing between install and vendored openshift/api #968
- Machine API Provider OpenStack Feature Gate #966
- OCPCLOUD-1252: Add validation webhook for guestAccelerators on GCP #927
- README: improve Controllers section indentation/naming #964
- Remove unused files #957
- Implement conditions for Machine Deletion Hooks reporting #946
- Azure Marketplace Machine Provider Bump #956
- Implement machine pre-drain and pre-terminate deletion hooks #945
- Update for golang1.17 #951
- GCP CI runs are complaining about APIs not being enabled #949
- Add powervs-machine-controllers to images configmap #948
- Rewrite rollout checks to requeue and not block reconciles #947
- Bug 2005052: Deny selector updates via webhook to prevent leaked machines #944
- Migrate to openshift/api #943
- Bug 1986237: add avg_over_time to the MachineNotYetDeleted alert #941
- Bug 2010368: fix mispelled field in alert rules #942
- Remove provider dependencies #938
- Fix node in MHC unit test #930
- PowerVS support #923
- Fix flaky AWS provider unit test #931
- remove kubevirt credentials request #940
- Bug 2013895: Bump Azure dependency to include new fields #935
- Update Dockerfile with correct image and go version #939
- Remove “kubevirt-machine-controllers” image permanently #936
- Bug 1986237: modify MachineNotYetDeleted alert to ignore pod #913
- Bug 2010368: modify alerts to contain summary and description #929
- Bug 2012228: fix(ibmcloud): Set resource group policy in CredentialsRequest #928
- Update OWNERS to use aliases and delegate MHC to dragonfly team #885
- podsecurity: enforce privileged for openshift-machine-api namespace #924
- Vsphere sessions check: fix log string formatting #922
- Add logging for possible error during vsphere session creation #920
- Bug 1999425: MaxUnhealthy should not be a string type #909
- Remove ironic-inspector reference in MAO #915
- docs/user/nodelink-controller: Fix name/namespace order for machine.openshift.io/machine #899
- Adds ibmcloud mapi controller to image ref #892
- Add service account names to gcp credentials request manifest #883
- Add the IBMCloud cluster api controller #871
- add a distinct status message for NoOp mode #914
- fix time values in alert doc #916
- Fix user-agent in vCenter sessions list #912
- Add webhook to ensure that only ID references AMI #888
- Full changelog
- Updating ose-mdns-publisher images to be consistent with ART #43
- Updating ose-mdns-publisher images to be consistent with ART #42
- Updating ose-mdns-publisher images to be consistent with ART #38
- Full changelog
- Adjusted test ns for the new ocp security model (#61) #61
- Updating ose-network-metrics-daemon images to be consistent with ART (#59) #59
- Updating ose-network-metrics-daemon images to be consistent with ART (#46) #46
- Updated prometheus client_golang version to v1.11.1 (#48) #48
- Fix field selector (#49) #49
- Fix golint to make the CI pass (#42) #42
- Fix prometheus label (#41) #41
- Full changelog
- bump kube to 1.24.4 #81
- bump kube to 1.23.1 #75
- Bug 2049155: fix printers to avoid panic from DeepCopy #71
- bump golang to 1.17 #65
- Bug 1986504: bump openshift/api #63
- registry: audit deleted access/authorize tokens #62
- Updating ose-oauth-apiserver images to be consistent with ART #59
- Full changelog
- Update logo #245
- fix json examples for –openshift-delegate-urls #203
- Use resourceName in Openshift SAR rule #243
- Updating golang-github-openshift-oauth-proxy images to be consistent with ART #241
- Bug 2026860: ocp provider: don’t fail client creation if oauth-server cert is not present #238
- Golang bump 1.17 #235
- Full changelog
- Updating oauth-server images to be consistent with ART #109
- Bump library-go to b2073c41366aae78868b24e86d99021b79d754cb #112
- fix linter findings #98
- Bug 2086505: Updating oauth-server images to be consistent with ART #108
- Bug 2086465: External OAuth: create audit logs for auth events #103
- AUTH-100: Add request header authn auditing #106
- Update external dependencies #105
- osin: Import from the new openshift module path #104
- AUTH-66: turn on audit on oauth-server with PW-based flow #92
- Bug 2038691: groupsync: create annotations if not present before writing to them #96
- Golang bump 1.17 #94
- Bug 2020625: oidc: allocate claims slice to 0 len to prevent empty groups #93
- Updating oauth-server images to be consistent with ART #84
- oidc handler: add remote OIDC URLs to error logs #74
- AUTH-8: Add group synchronization from OIDC providers #87
- Updating oauth-server images to be consistent with ART #83
- Full changelog
- OCPBUGS-2223: Update Default CatalogSource Images to 4.12 tag #487
- Updating marketplace-operator images to be consistent with ART #479
- OCPBUGS-441: Dockerfile.okd: include custom OperatorHub manifest #481
- (psa) make marketplace namespace restricted #484
- (namespace) remove PSA baseline enforcement #483
- Add PSA enforcement label #480
- Bug 2010375: Clarify reason/steps to diagnose in *CatalogError prom alert #469
- Bug 2088541: Add psa anotations to namespace to suppress warnings #478
- Updating marketplace-operator images to be consistent with ART #473
- Bug 2092464: Update default catalogsource to use 4.11 images #477
- Bug 2079610: Fix operatorHub status #470
- Revert “Bug 2092464: shift marketplace operator catalogs default to v4.11” #476
- Bug 2092464: shift marketplace operator catalogs default to v4.11 #474
- manifests/deployment: comply to restricted pod security level #471
- chore(modules): bump openshift/api for new operatorhub crd #472
- Bug 2057558: increase polling time of status report to reduce log spam #461
- fix(manifests): add singleton operatorhub cr #468
- fix(capabilities): add operatorhub crd manifest #467
- Bug 2070792: chore(manifests): add missing capabilities annotation #465
- manifests/10_clusteroperator: Drop the unused namespace property #401
- chore(manifests): add openshift capability annotation #462
- Bug 2067909: Bump github.com/prometheus/client_golang to v0.12.1 #463
- Update OWNERS #464
- Bump repository to Go 1.17 and k8s v0.23.x #460
- Updating marketplace-operator images to be consistent with ART #459
- Bug 1927478: Add catalog source pod spec overrides #457
- Bug 2034409: Update defaults to point to 4.10 images #456
- Bug 1996928: Revert “Start without defaults on ARM” #420
- Updating marketplace-operator images to be consistent with ART #428
- Update OWNERS #430
- Add summary and description for prometheus alerts #429
- Updating marketplace-operator images to be consistent with ART #427
- Remove the marketplace v1 and v2 APIs #426
- Remove unused packages #424
- Bump the k8s dependencies to 1.22.1 and controller-runtime to 0.10.0 #422
- Makefile,test: Fix flaky e2e tests #425
- Full changelog
- Update annotations to account for removal of duplicate whitespace #27532
- Set Jakub as a Test Platform team lead #27528
- k8s 1.25.0 #27526
- OCPBUGS-3282: Fix apigroup tag missing closing brace. #27525
- refactor how tests are run #27516
- Store monitor events in artifact directory at the end of run monitor #27515
- Add bond cni test #27405
- Add more resource to watch #27514
- only print the timeline into artifact dir to reduce noise #27509
- Write test failure summary, and add new command to request failure risk analysis from sippy. #27486
- upkeep: test data update #27501
- Revert 1.25 again #27499
- Extend SCC tests with their corresponding service account variants #27491
- MicroShift: label e2es with more apigroup labels (squashing more PRs into a single one + fixes) #27498
- Skip resource growth checks in upgrade-rollback jobs #27450
- k8s 1.25.0 #27490
- Remove must-gather pod from lack priorityclass exception list #27469
- Ocpbugs 2275 remove templateservicebroker test #27484
- Support external topology for hypershift #27437
- USHIFT-481 Get event intervals for alerts only if Prometheus route exists #27467
- Revert “k8s 1.25.0” #27489
- k8s 1.25.0 #27416
- Automated - Update synthetic test data #27475
- microshift: avoid oauth and console disruption monitoring #27462
- Migrate route.sh, run.sh, service.sh and status.sh tests to go #27455
- Add skips logic for tests that require optional capabilities #27481
- Fix flaky cli tests #27482
- Migrate idle.sh to go tests #27451
- Add dgoodwin as approver to origin #27477
- update skip openshift-controller-manager scaling events in build testsuite #27474
- Fix unit tests that have broken since we stopped running them 2 years ago #27460
- USHIFT-357: MicroShift: skip sig-scheduling tests #27402
- USHIFT-356: MicroShift: skip sig-node tests #27425
- OCPBUGS-2155: Revert “Add etcd vertical scaling test suite” #27463
- USHIFT-358: MicroShift: skip sig-storage tests #27438
- Check for console capability before launching it’s disruption monitoring. #27439
- Add Invariant test to look for probe errors for openshift-config-operator #27436
- Fix missing reporting of 0s disruptions. #27452
- Add etcd vertical scaling test suite #27444
- USHIFT-351: Microshift: skip sig-devex tests #27392
- USHIFT-300: microshift: skip sig-imageregistry tests #27367
- Drastically reduce the jobs we enforce disruption on. #27457
- USHIFT-354: MicroShift: skip sig-instrumentation tests #27399
- USHIFT-350: MicroShift: skip sig-coreos tests #27400
- egressip: avoid curl hanging when sending probes #27379
- OCPBUGS-1079: Use kubeconfig from secret mount instead of /tmp #27443
- include storage and must-gather tests in expectedTestCount #27356
- feat: update query results #27448
- USHIFT-283: microshift: skip sig-builds tests #27361
- Fix rare race condition in ImageStream admission tests. #27442
- remove test/extended OWNERS file #27408
- OCPBUGS-985: Revert “Test Baremetal deployment without a provisoning network” #27441
- USHIFT-344: MicroShift: fix sig-apps tests #27397
- USHIFT-301: microshift: skip sig-operator tests #27369
- USHIFT-343: MicroShift: skip sig-api-machinery tests #27410
- Remove test ‘clone repository using git:// protocol should clone using git:// if no proxy is configured’ #27391
- Bug 1933144: Bump k8s to include c83b5d076f5 #27380
- Check version when discovery client #27432
- USHIFT-285: MicroShift: allow CLI creation without OpenShift APIs #27358
- sig-api-machinery: remove project from authenticated browser APIServe… #27333
- Bump api-requests for console-operator on vsphere #27431
- USHIFT-353: MicroShift: skip sig-installer tests #27409
- make runonce-ldapsearch pod name unique #27426
- OCPBUGS-1265: test/extended/prometheus: Consider telemeterClient.enabled #27422
- Fix double disruption testing and values for most backends in intervals chart and bigquery. #27415
- add retry for checkLADPConn #27424
- e2e/cli: Migrate completions, whoami and projects tests to go #27371
- Enable ServiceInternalTrafficPolicy tests in OVNK #27043
- OCPBUGS-967: temporarily disable IngressClass test, which is changed in upstream in 1.25 #27417
- Add a test case to find job name/cluster version mismatches #27413
- upkeep: updated data set for 4.12 #27420
- OCPBUGS-1237: Fix s2i ruby test that relys on rack #27419
- Address exceeded maximum resolution of 11,000 points per timeseries #27384
- OCPBUGS-587: Temporarily flake FailedToUpdateEndpointSlices #27418
- Ignore the “InsufficientInstanceCapacity” patholocial event #27385
- Automatically skip tests with apigroups which are not served #27406
- Label tests with openshift crafted API groups #27387
- skip route-controller-manager scaling events in build testsuite #27394
- Remove flake for 40% Secret/ConfigMap growth test. #27381
- Bug 2100091: Ignore FailedCreatePodSandBox events for garbage collector tests #27270
- Fix upgrade dry-run. #27388
- Update etcd tests with k8s 1.25 data #27396
- Update etcd tests with k8s data #27393
- Instead of returning errors when there is no etcd operator, just return static values for etcd metadata #27342
- Test baremetal deployment without provisioning network #26676
- Ensures operators use external cloud provider #27349
- Bug 2098053: Re-revert “Add networking test for invalid external gateway” #27277
- Separate out pathological event test for error updating endpoint slices #27377
- Revert “Label tests with openshift crafted API groups (round 3)” #27378
- Skip etcd scaling test for metal and vsphere platforms #27375
- Test we don’t see substantial Secret/ConfigMap growth during upgrade. #27354
- Add ovnkube-node pods in the interval chart for ingress #27372
- update service load balancer test to gracefully roll out #27180
- EgressIP: Enable EgressIP tests for OpenStack #27363
- Label tests with openshift crafted API groups (round 3) #27368
- Suppress failures if certain firing alerts happened with related event #27304
- Bug 2095623: Revert “Disable azure-file tests” #27259
- job names: exclude text before
e2e for both platform and network tests #27373
- upkeep: updated synthetic tests historical data for alerts and disruption #27364
- Update job name tests for hard failure cases #27362
- Label tests with openshift crafted API groups (round 1) #27355
- Add required pod security admission annotations into cmd/namespace #27357
- Bug 2110503: EgressIP: Probes and check tcpdump logs should succeed eventually #27318
- test/extended/util/disruption: Remove stale p95 references #27351
- display UTC time for timeline charts #27348
- networking/acl: live-retrieve the NS instead of f.Namespace #27347
- Consider Prometheus maximum points in a time series for alert tests #27319
- Tune watch limits for operators #27281
- allow techpreviewnoupgrade alert when testing techpreview #27343
- Ignore repo name when looking for network type in job name #27339
- Use “plugin” for some nw isolation test case names #27330
- Revert “Revert ‘bump o/k for the fixed scheduling tests’”; layer fix on top #27332
- Update authenticated users RBAC e2e test #27326
- OWNERS_ALIASES: add me as a metal platform lead #27337
- Add Jan as image approver, remove Ravi and Clayton #27331
- Bug 2114506: test/extended/olm: Handle unset enabledCapabilities in marketplaceEnabled #27336
- Convert kubelet logs into monitor events for timelines #27299
- Revert “bump o/k for the fixed scheduling tests” #27328
- etcd vertical scaling test: add verbose timeout error messages and bump machine provisioning timeout for Azure #27322
- remove CLI.SetupNamespace() in favour of CLI.SetupProject() #27325
- Increase API requests limit for monitoring operator on openstack #27324
- bump o/k for the fixed scheduling tests #27323
- Separate out test for SamplesImagestreamImportFailing alarm #27314
- Fix same test name being used twice for pods on different nodes check. #27315
- etcd storage tests: work with dynamic namespace names #27295
- Add etcd vertical scaling test.” #27294
- Add a new spyglass chart for image-registry, console, oauth #27297
- Bug 2098194: set sampleErr to bodyReadErr #27260
- Fixes OVN ACL audit log parsing #27312
- Bug 2079679: pkg/monitor: wait for Prometheus sidecars to be ready #27224
- Add test that checks CI job name matches platform and network type #27311
- examples: drop OCP 3 library paths #27266
- Revert “Enable EgressIP tests for internal targets” #27307
- fix(olm-test): skip test if marketplace is not enabled #27302
- Bug 2089966: use multi-arch ldap and multicast test images #27169
- Revert “Bypass tests failing due to metal image rebuild on RHEL9” #27288
- Update to go 1.18 #27253
- separate interval gathering from interval calculation #27298
- Make margin wider to see more in spyglass charts #27061
- Bug 2098096: Enable EgressIP tests for internal targets #27262
- test/extended/prometheus: clean alert exceptions #27211
- test: allow -f to match tests for any test suite (redux) #27096
- BUG 2067323: test/extended/router: Drop host lookup for gRPC HTTP/2 h2spec tests #27274
- Bug 2097297: Show DNS lookup sampler problems on the disruption intervals chart. #27261
- Bug 2094342: bump cluster-capi-operator apirequests upper bound to 50 #27225
- Add Ravi and Abu as image approvers #27282
- Remove etcd vertical scaling test. #27280
- make a new category of pod sandbox failure #27247
- Bug 2099698: revert: no longer needed logic for rt kernel #27264
- Bypass tests failing due to metal image rebuild on RHEL9 #27283
- Bug 2098053: Re-revert “Add networking test for invalid external gateway” #27273
- Bug 2098053: Re-revert “Add networking test for invalid external gateway” #26985
- Bug 2100439: Enable GCP PD in-tree tests #27272
- Bug 2086519: AUTH-133: test/extended/security/scc: comply to pod security admission #27269
- Bug 2093122: synthetictests: Event struct’s Related field is optional #27208
- Bug 2086519: authorization: add a test checking that the restricted-v2 SCC mutates securityContext to match restricted PSa #27257
- Bug 2098095: EgressIP tests: Use OpenShift API to retrieve CloudPrivateIPConfigs and EgressIPs #27243
- Bug 2099611: Increase etcd-operator watch channels upperbound for BareMetal Platform #27263
- Bug 2098094: EgressIP test: Remove print statement on test failure #27191
- Bug 2093339: move provisioning test to permanent disablement, since this is not yet officially supported #27267
- Bug 2086092: Re-enable tests #27258
- Bug 2097713: Extend the allowance for etcdHighNumberOfLeaderChanges. #27238
- Bug 2097684: Nutanix provider for e2e testing #27102
- Add David Peraza as approver on samples test content #27241
- Bug 2093992: Bump timeout for cvo ugprade check to 10 minutes for Baremetal jobs #27254
- Bug 2086092: update kube to v1.24.0 #27244
- Bug 2097297: Don’t consider DNS lookup failure disruption #27250
- Bug 2093051: use the node update finished as a backup completion time for OSStaged #27206
- Bug 2086092: Extend time for kube rebase #27252
- Bug 2097009: Run must-gather tests in later to reduce resource contention #27249
- Bug 2086086: update samples for 4.11 #27234
- Bug 2085997: Report late requests verification: Fix for Hypershift #27220
- Bug 2094919: Skip etcd vertical scaling test on vsphere #27236
- Bug 2092961: Use http for network endpoint test. #27217
- Bug 2084361: Print error msg before os.Exit() #27230
- Bug 2093049: test/extended/networking: fix [pod sysctls should not affect node] to use known host interfaces #27221
- Bug 2085997: explicit timeout for the etcd scaling test #27216
- revert 1.24 tests to get back to work #27233
- Bug 2086092: update kube to v1.24.0 #27181
- Revert “Bug 2079679: pkg/monitor: wait for Prometheus sidecars to be ready” #27223
- Bug 2091110: Skip ErrImagePull for expected failures #27202
- allow terminating pods to be ready=false #27212
- Revert “Bug 2093049: test/extended/networking: fix [pod sysctls should not affect node] to use host interfaces” #27219
- Bug 2092961: Flake almost all ci cluster to external svc disruption. #27210
- Bug 2093049: test/extended/networking: fix [pod sysctls should not affect node] to use host interfaces #27203
- Bug 2093234: Add regex filter and intervals duration to e2e-charts #27186
- Bug 2093046: priority class: add exception for must-gather pending bug fix #27204
- Bug 2081732: remove incorrect namespace check in static pod test #27195
- Bug 2092961: Add disruption test to an external service, not cluster under test. #27179
- Bug 2091547: Skip internet connection test if we use a proxy #27093
- Bug 2079679: pkg/monitor: wait for Prometheus sidecars to be ready #27201
- Bug 2091067: EgressIP: Make tests retry on API timeout errors #27185
- Bug 2085997: report late request (broken LB) for the external client #27129
- Bug 2085997: brings back the etcd scaling test #27176
- Bug 2091086: test/extended/cli/explain: Include OperatorHub CRD when ‘marketplace’ is enabled #27188
- rename intervals to timelines to be self-explanatory #27182
- Bug 2086519: label project-created namespaces to not sync PodSecurity labels #27177
- wait for kube-apiserver to settle after modifying network configuration #27193
- Bug 1992596: move help.sh to golang-based e2e #27190
- Bug 2091086: cli: don’t insist on metal3 CRDs being present when cluster capabilities are None #26998
- restrict the per-namespace invariant alert check #27184
- Bug 2090692: slightly loosen oc exec help test #27183
- Updating openshift-enterprise-tests images to be consistent with ART #27139
- allow up to two DNS check pods to be waiting #27175
- Bug 2086092: skip tests to facilitate rebase #27178
- Bug 2081732: Static pod test error #27160
- Revert “Bug 2089831: Temporarily disable telemetry token test.” #27168
- add per-namespace tests for kubepodnotready alert failures #27151
- remove the etcd vertical scaling test #27174
- Bug 2089276: Detect and use the packet capture interface for EgessIP tests. #27164
- remove test: ‘etcd is able to vertically scale up and down with a single node’ #27170
- Bug 2086519: bump o/kube so that test namespaces aren’t PSa label synced #27173
- Bug 2085997: special configuration on barematal platform for the etcd vertical test is required #27140
- don’t trigger single second disruption on contiguous disruption #27161
- Bug 2089831: Temporarily disable telemetry token test. #27165
- BZ2088533: Add subresource.status check exceptions for techpreview jobs #27156
- add pod graceful termination to pod lifecycle #27153
- Bug 2081732: Dump debug message whenever static pod test fails #27143
- Bug 2086086: image_ecosystem: remove deprecated imagestreams #27133
- Bug 2086519: Auth-133: test/extended/util: comply exec pod to restricted pod security profile #27152
- Bug 2085997: makes the etcd invariant test work on dualstack platform #27148
- Bug 2085997: ExternalControlPlaneTopology: Skip etcd master endpoint verification #27149
- BZ2087992: Add regex matching, regex inverse matching, and default end date option for intervals #27117
- BZ 2087504: Add Early test to check for CRD subresource.status #26954
- Bug 2085997: the machine API can be unavailable resulting in a 404 or an empty list #27147
- EgressIP downstream tests for OVN Kubernetes #26999
- Add user coreydaley as an approver #27144
- handle cases where containers restart #27137
- commenting ipv6 for sysctls #27142
- Bug 2085997: ensure number of voting members in the cm matches the number of master machines #27141
- Bug 2085997: the etcd vertical scaling feature misses e2e tests #27127
- Bug 2084361: Add some debugging information for mysterious exit 2 errors #27136
- Bug 2085997: adds new test to the API monitor #27118
- update alert and disruption limits #27130
- Bug 2085997: gives the backend monitoring an identity #27134
- always allow at least one second for disruption #27131
- Bug 2080007: unrevert external ip configuration e2e networking tests #27078
- Bug 2083941: feat: add fuzzy match on micro release #27123
- add test to catch single second disruption patterns #27132
- Bug 2082538: bump cluster-capi-operator apirequests upper bound for GCP And Azure #27108
- Revert “moves TestScalingUpAndDownSingleNode from cluster-etcd-operator” #27126
- Add debugging message for static pod test #27111
- Bump net.ipv6.neigh.IFNAME.retrans_time_ms for Multiarch #27119
- Don’t hardcode the SA token audience when requesting one #27120
- timeline command, container exit fix, key resources by uid, and use reflector #27114
- NE-883: test/extended/router: Add test for spec.subdomain #27030
- moves TestScalingUpAndDownSingleNode from cluster-etcd-operator #26917
- Add config for golangci-lint no-sprintf-host-port #27116
- Correct pod timing #27112
- Bug 2080687: request prometheus service account token #27100
- Fix request token ic cli/admin tests #27113
- Bug 2081012: Don’t wait for sa token #27104
- prometheus helper: add helper to retrieve prometheus query token #27107
- Bug 2080679: fix oc related test to land k8s 1.24 #27099
- [rebase 1.24] disable tests with time limit to facilitate rebase #27086
- Bug 2082239: Bump net.ipv6.neigh.IFNAME from 30005 to 30010 for Multiarch #27101
- test/extended/util/openshift/clusterversionoperator/adminack: Accept MultipleReasons #27110
- make it easy to create pod interval tests #27092
- Bug 2080681: Request new token instead of using node-bootstrapper token #27098
- Revert “NE-626: Add DNS CI coverage for golang and glibc resolver libraries” #27103
- make the pod ip check a controller #27097
- NE-626: Add DNS CI coverage for golang and glibc resolver libraries #26957
- Revert “test: allow -f to match tests for any test suite” #27095
- MON-2148: Handle 401 code for prometheus api route #27031
- test: allow -f to match tests for any test suite #27091
- router: wait for “use scc restricted” SAR before pod creation #27084
- Remove non-functional gu workload tests. #27044
- test/extended/prometheus/image_pulls: New “should be fast” test-case #27067
- give multi-arch more time to complete upgrades to clean up signal #27085
- skip ‘should not see excessive Back-off restarting failed containers’ for e2e tests #27087
- haproxy needs to request privilege escalation in order to work #27080
- updates “managed cluster should report ready nodes the entire duration of the test run” #27077
- skip etcd quorum probes failures when etcd revision changes #27072
- don’t wait for SA token secrets #27081
- Revert “Add e2e tests for ExternalIP Policy and AutoAssignCIDRs” #27075
- Add restricted-v2 to the default RBAC rules for authenticated users #27071
- Increase determinism when searching for prom service secret token #27073
- improves etcdRevisionChangeAllowance calculates the number of etcd’s operator rollouts #27070
- Include timestamp when reporting observed ReusedPodIP events. #27074
- track pod IP re-use #27062
- hack/verify-generated: Replace –quiet with –exit-code #27060
- Bug 1992596: e2e/cli: Migrate create.sh to Go tests #27049
- force completes upgrade failure if operator failure #27058
- Bug 2078528: bump cluster-node-tuning-operator apirequests upper bound for OpenStack #27052
- handle cases where the DNS querier pod isn’t yet running #27056
- Add e2e tests for ExternalIP Policy and AutoAssignCIDRs #26894
- tuning-cni: basic tests #26953
- Sysctl: validate that setting a pod sysctl does not affect node and other pods #26940
- the ‘should be scheduled on different nodes’ were only fixed in 4.11 and later #27053
- Separate out ‘Back-off restarting failed container’ from pathological test #27047
- Revert “Skip newly enabled networkpolicy tests on IPv6” #27045
- Flake pod sandbox errors pinging container registry on azure. #26964
- Ignore loki NetworkNotReady repeating events. #27051
- Bug 2030972: test/extended/util/openshift/clusterversionoperator: Survive API hiccups #27041
- Bug 2077457: Skip BeforeEach as well for the skipped HAProxy config manager test #27034
- synthetic: cloud api quota synthetic test should always be present #27042
- Remove fixed clusterIP service for ipv6 contexts #26895
- loosen SCC capabilities check to allow stricter caps #27024
- test/dns: add the bits to make the Restricted PodSecurityPolicy happy #27035
- Disable FIPS disabled serving routes test on BM #27033
- test/e2e/upgrade: Use PATCH instead of POST for ClusterVersion spec updates #27023
- Sysctl tests: pod with sysctls not on whitelist should not start #26937
- Bug 2075584: Use substring match instead of exact match #27028
- Switch to groupified APIs #27025
- Dump imagestreams and buildconfigs on build imagechange trigger test failure #27019
- Best matcher single node OVN #26929
- : Separate pod network creation failure from others in sandbox creation… #27020
- test/…/disruption: set pod security level centrally #27017
- Switch to groupified APIs #27008
- router: construct url with net.JoinHostPort #27015
- Add exception for etcd guard pod readiness probe error events #27016
- test/extended/*: set necessary pod security settings #27014
- getArchitecture: Fix to work with Hypershift #27004
- Bug 2072171: Reenable the implementation for basic egressfw e2e CI test #26973
- MGMT-9440: Fix single node serial tests API crash #26904
- skip tests for etcd leaders when etcd revisions change a lot #27009
- Fix another issue with CVO upgrade ack timeout on metal. #27001
- bump o/kube to get latest PSa test updates #27012
- Bug 2031564: Separate out test for ‘RequiredInstallerResourcesMissing secrets’ #26936
- Bug 2060406: bump apirequests upper bounds for GCP #27002
- Bug 2072533: synthetictests: skip “alert/Watchdog must have no gaps or changes” on SNO upgrades #26976
- Bug 2074445: exclude loki-promtail from duplicated events #27006
- Fix up one more optional capabilities new-app test #27005
- actually set default PSa labels if unset #27000
- util/client.go: add method to specify pod security admission level #26938
- Add test for validating related object references. #26921
- correct pathological event detection #26994
- prevent npe for old container status readiness #26996
- Bug 2011895: Add synthetic test for cloud API throttling #26559
- router: fix all uses of Sprintf to build host:port for a URL #26983
- prevent pending/firing alert overlap #26968
- Skip newly enabled networkpolicy tests on IPv6 #26977
- test: switch to testing CapBnd over CapInh #26960
- remove wait for samples imagestreams, no longer needed #26992
- Bug 2066865: bump openshift/kubernetes #26969
- test/extended: add/update OWNERS files for Prometheus #26910
- test/extended/prometheus: Remove SDN/OVN-K alert rules #26687
- Unrevert and fix for the HAProxy test #26980
- Bump timeout for CVO to ack the upgrade generation. #26974
- Bug 2066457: don’t fail when a query returns warnings #26984
- Bug 2072924: increase the threshold on the number of telemetry series #26988
- Update limits #26982
- fix evaluation of deleted pods for pod sandbox test #26978
- Revert “Fix HAProxy tests on FIPS properly” #26979
- Revert “Add networking test for invalid external gateway” #26975
- Revert “Implementation for basic egressfw e2e CI test” #26967
- Bug 2067323: extended/test/router: omit DNS managed conditions for shards #26949
- Add networking test for invalid external gateway #26883
- Bug 2060406: test: bump apirequests upper bounds for VSphere and AWS single node #26962
- test: update synthetic test node restart count #26948
- bump: k8s.io/kubernetes #26955
- Ignore FailedCreatePodSandBox event that is expected to happen #26966
- add pod lifecycle intervals to separate pages #26908
- Add missing operators from the known operators list #26963
- Bug 2047913: Fix HAProxy tests on FIPS properly #26803
- Bump allowed watches for node-tuning-operator #26944
- Migrate tests away from using samples-operator provided imagestreams #26913
- Image registry redirect #26920
- Bug 1996883: increasing build config loglevel #26958
- Use ServerGroupResources instead deprecated ServerResources #26956
- Fix typo in OVN Kubernetes skipper #26950
- Bug 1996883: Increasing log level to support decrease in default buildah log level #26941
- Implementation for basic egressfw e2e CI test #26934
- Check token response error code #26926
- Clarify that the image mirroring requirement applies to conformance tests only #26912
- coarse serialization of storage tests #26933
- Revert “bump: k8s.io/kubernetes” #26939
- bump: k8s.io/kubernetes #26935
- Flake on FailedCreatePodSandbox if missing ovn readiness file. #26932
- Add pod securityContext sysctl whitelist tests #26809
- Revert “Implementation for basic egressfw e2e CI test” #26931
- Update where the canary test lists is retrieved from #26930
- Implementation for basic egressfw e2e CI test #26865
- Implementation for egress-router-cni e2e ci test #26875
- Attempted fix for missing monitor events. #26862
- JKNS-267: minimize jenkins e2es (pipeline strategy verification only) #26914
- Attempt multiple different token requests #26916
- Simplify 26907 #26924
- Bug 2034688: allow Prometheus/Thanos to return 401 or 403 when the request isn’t authenticated #26695
- Migrate builds tests away from using sample imagestreams since samples will be optional #26909
- Update prometheus owners #26911
- Pod scheduling tests improve pod deployment identification #26906
- Add early test to check router pods are not on same node #26899
- Give each interval chart a title #26896
- Mark image pull from registry.redhat.io as flake to track but not fail #26897
- Bug 1961233: Add DNS availability upgrade test #26795
- Revert “Disable intree vsphere tests” #26797
- Bug 2026063: build-quota test: fix cgroupv2 parsing holes exposed by buildah upgrade #26885
- Bug 2057990: Add debug info for signature test #26884
- MGMT-9440: Fix single node serial tests API crash. #26868
- Bug 2060498: only add failure when no event was found for the target revision #26880
- Bug 2060406: operators should not create watch channels very often: bump OpenStack monitoring operator #26877
- Use common threshold constant as other event tests #26873
- cleanup network policy ACL extended test #26876
- add text to the timelines when you hover #26871
- Make a test to specifically look for image pull backoff for registry.redhat.io #26858
- Bug 2059299: IBMCloud: Ignore CRB already exists #26864
- Switch to batch/v1 CronJob #26867
- Bug 2059277: images/tests: yum update python3-six to workaround ART issue #26863
- Increase scale test timeout to 30 minutes #26861
- Fix ovnkube-node readiness test to include successes. #26860
- Bz2054254 rearrange test cases #26857
- Set ovnkube-node readiness failure to flake for now. #26855
- Increase cluster-monitor-operator watch count threshold on single-node clusters #26685
- allow slow kubelet updates for static pods, but still fail if they never succeed #26852
- OWNERS: Declare Bugzilla component #26368
- vsphere is firing this alert/VSphereOpenshiftNodeHealthFail and it is hiding other alerts #26850
- Updating openshift-enterprise-tests images to be consistent with ART #26804
- update alert thresholds after extending static pod install duration #26844
- ip reconciler ignore #26842
- Bug 2049117: Reenable wait on worker deletion and increase serial test timeout #26810
- Bug 2053582: fail on static pod lifecycle failures #26837
- Allow a single test to override suite timeout #26833
- Bug 2053312: requestheader test must wait for authentication operator to settle after config update #26834
- Bug 2053143: allow inexact matches for disruption data #26831
- Break out test for OSUpdateStaged event with no OSUpdateStarted. #26824
- add job type key to the error we track for statistical jobs #26823
- baremetal: add test for preprovisioning images #26780
- Bug 2045590: Alibaba: enable init for alibabacloud provider #26818
- Revert “Flake failed sandboxes from bug in new guard pods” #26817
- Debug missing OSUpdateStarted events #26798
- Delete extra workers before running metal high availability test #26813
- images: update registry to 2.8.0-beta.1 #26811
- Bug 2044824: Update k8s vendoring #26805
- Bug 2050345: update p99 values for disruption and alerts #26815
- Bug 2047911: IBMCloud: Concurrent CRB create #26806
- Only run parallel tests in the canary job since it runs in parallel #26812
- Fix broken console disruption test. #26808
- Add baremetal high availability tests #26569
- Bug 2047362: Prometheus check targets endpoint #26793
- Verify Builds with CSI Volume Sources #26791
- Bug 2047790: Skip some HAProxy tests on FIPS #26800
- don’t double report early alert failures #26796
- add more frequently failing loki alerts to the allow-list #26788
- Bug 2039539: Allow apiserver burn rate alert to fire in CI #26748
- Revert “add support for build csi volume source” #26790
- Flake instead of fail the timely OSUpdateStaged test on metal and ovirt. #26789
- Bug 2043042: Additional checks for requestheaders.go #26777
- Bug 2045590: Adding new cloud provider e2e for alibabacloud #26782
- use real alert data to determine flake and failure timeframes for alerts #26781
- Bug 2040793: Fix ordering issues with snapshot e2e #26778
- Bug 2044210: [baremetal] remove wait condition when deleting extraworkers #26774
- Bug 2034285: test/extended: re-enable API data in etcd test #26768
- Bug 2038481: Flake failed sandboxes from bug in new guard pods #26776
- adds a synthetic test for checking if the pod.spec.nodeName field is immutable #26772
- storage operator can go upgradeable=unknown if vsphere is unreachable #26773
- Bug 2035534: Baremetal skip inspection #26681
- Bug 2037075: Verify Builds with CSI Volume Sources #26646
- allow loki target to be down #26771
- Bug 2033720: sync imagestreams #26750
- loki moved namespaces, so we have to move to match #26769
- Bug 2005901: Allow KA guard probe to fail as designed #26766
- update disruption times and shift to p99 #26767
- Bug 2033751: Rebase 1.23.0 #26753
- Add test for OS updates being staged within 5-10m. #26759
- Bug 1931594: add one more timeout condition for –request-timeout test #26756
- separate prometheus watch alerts #26762
- Alter junit test suite names for jobs with multiple junit xmls. #26755
- split watchdog test into its own test #26760
- allow promtail restarts as designed #26761
- make it easy to create multiple flavors of interval charts #26757
- Bug 2040741: Allow baremetal tests to run on multiple platforms #26660
- Bug 2038389: Fix vsphere testing workarounds #26737
- allow burn budget to flake until kube-apiserver team finds and fixes it #26747
- for upgrades, skip alerts with individual invariant tests #26746
- Display ErrImagePull messages. #26745
- Record zero for known alerts #26744
- stop flaking on disruption because we always produce data for analysis #26735
- separate out more alerts into specific tests #26743
- watch specific alerts for pending and firing #26741
- allow wardle aggregated apiservers to be down pending bug #26742
- use route to access prometheus instead of exec pods. #26740
- Bug 2039074: Revert “Add workaround for baremetal resources test” #26734
- Bug 1981867: add backoff retries for oc explain tests #26728
- test/extended/util/annotate/generated: Delegate to team leads #26714
- Round before comparing disruption #26731
- making missing p95 obvious to see and add topology to our search keys #26730
- Add tests for ErrImagePull monitor intervals. #26702
- update p95 values #26726
- Bug 2033720: Drop usage of old imagestreams #26708
- Bug 2026488: Skip Duplicate openshift-controller-manager Events #26719
- allow skipping a backend test #26721
- set allowed disruption based on historical p95 data #26717
- Bug 2035128: disruption: use proxy from environment when measuring disruption #26720
- add downtime checks for console and oauth-server #26707
- Bug 2033751: Disable etcd test for 1.23 #26712
- Bug 1983758: Add GCE back into the frontend disruption test #26699
- Bug 2033810: pkg: Buffer signals for abortCh #26715
- extend timeouts with 5s for each part #26713
- find disruption backends from the event intervals instead of a hardcoded list #26706
- Bug 2033740: Implement e2e-images-update label #26692
- extend tcp dial timeout and http response header timeout #26709
- use standard disruption checker for load balancer #26703
- oc label will properly report removed label #26700
- fix disruption sampler #26682
- Add workaround for baremetal resources test #26693
- test: avoid double slash in path #26691
- prometheus: adjust telemetry test for remote_write switch #26631
- Revert “Alter junit test suite names for jobs with multiple junit xmls.” #26686
- test/extended/util/disruption: Include tolerated percentage when failing #26653
- force a new payload #26683
- Fix TestFIPS on Hypershift #26677
- Hypershift: Skip Prometheus should have important platform topology metrics #26678
- Skip must-gather tests when running on Hypershift #26648
- test/extended/etcd: cleanup OWNERS #26673
- ensure that we always produce output for successful pod sandbox tests #26674
- Only known images used by tests: Fix for Hypershift #26641
- Adjust build test images for multi-arch support #26598
- Disable intree vsphere tests #26670
- Fix etcd leader changes are not excessive for Hypershift #26658
- OLM: Fix tests that check on pods when external ControlplaneToplogy #26632
- Revert “refactor disruption suite to collapse to a single implementation” #26671
- refactor disruption suite to collapse to a single implementation #26669
- Alter junit test suite names for jobs with multiple junit xmls. #26667
- Bug 2026806: clusterversionoperator/adminack.go: Check for nil cm map #26668
- Updating openshift-enterprise-tests images to be consistent with ART #26471
- Bug 2026806: Admin ack ignore unrelated upgradeable false #26661
- vendor: update openshift/kubernetes to include gcp quota fix #26663
- Revert “Add test to check kubernetes version” #26659
- OLM should have imagePullPolicy: Fix when ExternalControlplane #26638
- Bug 2026806: test/e2e/upgrade/adminack: Poll gates for duration of update #26649
- Skip oc version test in disconnected #26655
- test/extended/util/openshift/clusterversionoperator: Delegate admin-ack ownership #26650
- Move shared OWNERS file into generated dir #26652
- Add test to check kubernetes version #26595
- Specify default namespace when getting API endpoint #26616
- Update imagestreams in build and image-ecosystem #26612
- Fix ‘should only include cluster daemonsets that have maxUnavailable’ for Hypershift #26639
- double space preventing proper reporting for disruption #26642
- Fix ‘oc expose can ensure the expose command’ for custom service cidrs #26637
- OC explain should contain proper spec+status for CRDs: Skip some resources when controlplane is external #26636
- Local kubeconfig tests: Allow no master nodes when external controlpane topology #26623
- Bug 1880738: Minimise test disruption by removing newest workers in scale down #26469
- cli: use ProxyFromEnvironment for requests #26630
- Bug 1992596: e2e/cli: Migrate apiresources.go and add version #26589
- Allow even greater downtime on Azure single-node upgrade #26624
- Disable app behind service lb with pdb disruption test on single-node #26628
- Allow disruption in image-registry availability during single node upgrades #26625
- Allow operator-lifecycle-manager-packageserver to not define related objects when external controlplane #26627
- Bug 1992596: e2e/cli: Finish basicresources.sh migration #26620
- Bug 2023238: Skipping Django Test until bug is fixed #26617
- Bug 2015624: Refactor IBM Cloud skips to External #26537
- Skip imageregistry serial test for disconnected environments #26547
- Bug 2023631: Create new-project without updating kubeconfig #26610
- Bug 2022797: Switch the cluster-role-reapers test to Serial #26603
- olm: enable test on arm64 #26609
- Managed cluster should ensure control plane operator: Allow hypershift kapi proxy #26579
- kubeconfigs: fix capture loop variable in closure #26552
- Add parallel test for new baremetal resources #26541
- prevent rare alert gathering panic #26605
- Update baremetal tests owners #26608
- Added 2 Whereabouts e2e tests #26514
- pkg/synthetictests: add toleration for missing versioned etcd-endpoints cm during upgrade #26602
- Fix single-node frontends disruption test tolerated disruption % #26592
- Bug 2021141: Cluster should allow a fast rollout of kube-apiserver is… #26576
- account for longer tests runs in the watch pathological watch test case #26601
- Revert “Finish basicresources.sh migration” #26597
- oauth/requestheaders: escape semicolon in URI #26596
- oauth e2e: fix the expected http2 error to match golang 1.17+ #26594
- Refresh prometheus OWNERS #26591
- Set upper bounds to 50 percentile #26583
- Bug 2002776: fix intermittent test failure due to missing tools image #26580
- Bug 1992596: e2e/cli: Finish basicresources.sh migration #26578
- add alerts to tracked metadata #26571
- Bug 2021433: Disable new build test on disconnected #26581
- CLI tests: Softcode the kubernetes.default cluster IP #26574
- correct string matching for backend disruption summary gathering #26577
- Bug 1992596: e2e/cli: Add env tests #26573
- write disruption data #26548
- Bug 1992596: e2e/cli: Add probe tests #26570
- Disable topology tests for vSphere in-tree storage plugin #26565
- Bug 2011293: add default search image registry test to build e2es #26530
- Bug 2009879: Add admin ack Upgradeable condition gate test #26525
- Add baremetal operator validating webhook healthy test #26484
- Bug 2019375: tests: skip some tests for proxy #26558
- Bug 2000216: Add more tests for image policy #26536
- Bug 1999561: Extend API rollout resiliency test to validate pods restart during the rollout #26451
- Bug 1826225: test/extended/router: Enable edge-terminated gRPC #26456
- Oauth server tests: Get Oauth issuer url from discovery instead of route #26555
- Bug 2019516: Skip test ‘clone repository using git:// protocol should clone using git:// if no proxy is configured’ #26560
- Disable vSphere-specific storage tests #26554
- Update watch thresholds for cluster-monitoring-operator #26550
- Set provider name when platform is vSphere #26551
- Add myself to OWNERS #26544
- Add exception for repeated api check events on single-node #26543
- Fixup pod request/limits check when workload partitioning is enabled #26483
- Bump apirequestcount upper bounds for 4.10 #26538
- ignore ClusterNotUpgradeable alert on a TechPreviewNoUpgrade cluster #26532
- Ignore upgrade errors on vSphere HW 13 platformsupported #26533
- tigthen the SCC test for fixed levels, relax for unfixed levels #26540
- RBAC default rules test: allow new configmap to authenticated users #26524
- test/extended/operators: Allow early KubeletMinorVersionUnsupportedNextUpgrade #26531
- test/extended/operators: New staff-engineer OWNERS #26534
- move olm tests, owners to separate context #26500
- Add tolerance value to debug output #26520
- Bug 2011924: Fix inaccurate test error for NodeReady multiple times. #26527
- update canary tests for 4.10 #26528
- ignore ClusterNotUpgradeable alert on a TechPreviewNoUpgrade cluster #26515
- Bug 1960674: images: port image signature workflow test to OCP4/UBI8 #26510
- test/extended/prometheus: Validate alerting rules #26504
- Add another failing unidling test for ovn-kubernetes #26509
- create event intervals for alerts #26508
- operators should not create watch channels very often: be more verbose when failing the expected condition #26512
- Update test-cmd after recent removals in oc #26513
- e2e: Wait for an individual imagestream import #26472
- Bug 2007850: Revert “images: port image signature workflow test to OCP4/UBcI8” #26485
- Bug 1992596: e2e/cli: More basicresources.sh #26502
- Tolerate API downtime in upgrade disruption tests on single-node topologies #26498
- Skipping test “[sig-devex][Feature:ImageEcosystem][perl][Slow] hot deploy for openshift perl image hot deploy should work” for ARM64 #26507
- suppress NetworkPodsCrashLooping while we work out whether its a bug … #26506
- fix e2e namespace event typo #26505
- improve RestrictedEndpointsAdmission test #26423
- Bug 2003239: builds: skip private git repo tests outside of CI #26461
- record only known images test pass so we can get a percentage of passes #26494
- split the event should not repeat pathologically into e2e and core #26495
- Revert “stop adding individual serviceaccount namespace groups to SCC… #26501
- Revert “test/extended/prometheus: Validate alerting rules” #26499
- This would make CI for 4.10 in vSphere environment wee bit happier #26493
- Bug 2000754: vendor: e2e iperf2 change threshold to 10MBps #26491
- stop adding individual serviceaccount namespace groups to SCC for e2e tests #26487
- test/extended/prometheus: Validate alerting rules #26476
- Bug 1991980: Skipped non-standard openshift installed crd testing for ibmcloud #26388
- Bug 2008926: etcd-storage-test: mark node ready to save it for 5min from cloud provider #26490
- add disable-scc label to kube e2e tests #26486
- chore: ensure all etcd packages are using the same version #26489
- Remove exception for ovirt resource requests #26488
- Bug 1979962: Fix aws network stress tests #26359
- pkg/synthetictests: add etcd client connection closing to known events #26475
- Use PlatformStatus Type instead deprecated Platform #26481
- track the final state of pods and events from e2e tests for later debugging #26390
- Bug 1992596: e2e/cli: Move annotations.sh and basicresources.sh to proper e2e #26415
- Bug 2004009: bump(k8s.io/*): v1.22.1 #26466
- disable flaky unidling test #26467
- tests/extended/builds: handle new step logging #26464
- Bug 1998029: CustomNoUpgrade should not fire no-upgrade failures #26449
- Bug 1960674: images: port image signature workflow test to OCP4/UBI8 #26459
- Bug 1997509: delete only necessary resources instead of all #26442
- Bug 2003252: builds: fix BUILD_LOGLEVEL tests for nightlies #26463
- Run k8s test separately from OpenShift tests #26443
- Split the “Cluster frontend ingress” test into 4 #26445
- test: skip tests that won’t work behind a proxy automatically #26398
- test/extended[auth*,oauth]: configure OWNERS #26453
- Added E2E test for multus-cni #26347
- Bug 2001441: must-gather: Ignore startup logs in kube-apiserver audit logs #26450
- Full changelog
Source code for this page located on github