Description of problem: the project is deleted, mark as Terminating but never deleted. There is no info in controllers logs showing why the project can't be removed. customer has created project 'project-term' and deleted it, however, it shows only this: I0604 19:29:26.243530 1 vnids.go:115] Allocated netid 13363969 for namespace "project-term" E0604 19:29:27.103393 1 namespace_scc_allocation_controller.go:335] error syncing namespace, it will be retried: Operation cannot be fulfilled on namespaces "project-term": the object has been modified; please apply your changes to the latest version and try again E0604 19:29:28.192598 1 namespace_scc_allocation_controller.go:335] error syncing namespace, it will be retried: Operation cannot be fulfilled on namespaces "project-term": the object has been modified; please apply your changes to the latest version and try again There is finalizer: openshift in the project, but after removing that the project is still not removed This issue is for every project they have in the cluster. All remaining in status Terminating. Version-Release number of selected component (if applicable): OpenShift Container Platform 3.11.59 atomic-openshift-3.11.59-1.git.0.7cb6277.el7.x86_64 How reproducible: n/a create and delete project, it stuck in Terminating status Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: will attach the logs from master-api and master-controllers asap.
Please refer to https://bugzilla.redhat.com/show_bug.cgi?id=1703521#c6 which again refers to https://bugzilla.redhat.com/show_bug.cgi?id=1625194#c25 to see if caused by other aggregated apiserver.
Hello Xingxing, what would be the fix as it is not clear from the bugzilla: - run "oc get apiservices | grep False" and open bugzilla about the component? - deleting the "v1beta1.servicecatalog.k8s.io" apiservice - it is still needed? Thx
Then I have no idea, need help from Dev colleague. @Michal, could you help check when having time?
Saw you "controllers logs", it contains 'unable to register configz...register config "componentconfig" twice', which looks similar to bug 1689000 (see its 11th and 16th comments): I0605 13:28:30.627181 1 controller_manager.go:35] Starting controllers on 0.0.0.0:8444 (v3.11.59) I0605 13:28:30.649975 1 start_kube_scheduler.go:62] `kube-scheduler [--leader-elect=true ...snipped... ... I0605 13:28:30.696360 1 standalone_apiserver.go:101] Started health checks at 0.0.0.0:8444 I0605 13:28:30.722406 1 leaderelection.go:185] attempting to acquire leader lease kube-system/openshift-master-controllers... ... I0605 13:28:30.866861 1 start_kube_controller_manager.go:87] `kube-controller-manager [--kube-api-content-type=application/vnd.kubernetes.protobuf ...snipped... ... E0605 13:28:30.930723 1 controllermanager.go:134] unable to register configz: &{%!c(string=register config "componentconfig" twice)} I0605 13:28:30.930844 1 leaderelection.go:185] attempting to acquire leader lease kube-system/kube-controller-manager... E0605 13:28:31.184642 1 reflector.go:136] k8s.io/client-go/informers/factory.go:130: Failed to list *v1beta1.PodDisruptionBudget: Get https://bdb-pr-vm-master-2:443/apis/policy/v1beta1/poddisruptionbudgets?limit=500&resourceVersion=0: dial tcp 10.88.48.8:443: connect: connection refused Not sure if this causes the terminating issue. @Michal, any idea? Thanks.
I’m adding UpcomingSprint, because I was occupied by fixing bugs with higher priority/severity, developing new features with higher priority, or developing new features to improve stability at a macro level. I will revisit this bug next sprint.
This issue was assigned to me just yesterday. It seems it hasn't been updated for a while and all linked customer cases were closed. https://access.redhat.com/support/cases/02408525 and https://access.redhat.com/support/cases/02570051 indicate that the root cause of the issue was a custom finalizer on an object. Thus I'm closing this issue, feel free to reopen.
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 120 days