4.9.0-0.okd-2023-02-11-180223

New features since version 4.8.0-0.okd-2023-02-23-164510

Download installer and client with:

oc adm release extract --tools registry.ci.openshift.org/origin/release:4.9.0-0.okd-2023-02-11-180223

No tests for this release

Loading changelog, this may take a while ...

Changes from 4.9.0-0.okd-2023-03-29-123747

Created: 2023-02-11 18:03:23 +0000 UTC

Image Digest: sha256:0107313617814434848855346ef7351f6a09629790df573357f26461f65038cd

Components

Kubernetes upgraded from 1.22.17 to 1.22.15
Fedora CoreOS 49.34.7

Rebuilt images without code change

aws-machine-controllers git 0f5c46e8 sha256:9bab03a458e61f0315b70e26e009b410be6481b2b1f88031fd4b7b90f0a560cd
azure-disk-csi-driver-operator git 15e474ce sha256:bb2f15a69ab5b12d4df0ce0e1f81fb8e4f916a886ce0f501b9e26cde6f9b7847
baremetal-operator git 4e7605ba sha256:f10eb3c213febc0d39f3cadb221b9462f17744f6daab85fcb69ffe4efe6e0d95
keepalived-ipfailover git b3796438 sha256:4525e6a8617528e6a1dab9c1d355a6eadb6a77173ca0deadb2374c9466ee952d

tests

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

View changelog in Markdown or change previous release:

Source code for this page located on github