Bug 1609731
| Summary: | Create etcd-cluster in admin console got unexpected 404 Error | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | ge liu <geliu> |
| Component: | Management Console | Assignee: | Alec Merdler <amerdler> |
| Status: | CLOSED ERRATA | QA Contact: | Yadan Pei <yapei> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 3.11.0 | CC: | amerdler, aos-bugs, ecordell, geliu, htariq, jhadvig, jiazha, jokerman, juzhao, mmccomas, spadgett, ssadeghi |
| Target Milestone: | --- | ||
| Target Release: | 4.1.0 | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2019-06-04 10:40:22 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
ge liu
2018-07-30 09:54:06 UTC
Can you try with a more recent OCP release? I can't reproduce off of master. @ecordell, I tried with latest ocp version, still reproduce this error:
# openshift version
openshift v3.11.0-0.20.0
# oc get pods etcd-operator-7b49974f5b-dmsnz -o yaml|grep image
image: quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
imagePullPolicy: IfNotPresent
image: quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
imagePullPolicy: IfNotPresent
image: quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
imagePullPolicy: IfNotPresent
imagePullSecrets:
image: sha256:10639c807879e82aa694ca1189c8fd06844fcdc7120ecf15cc367021bd392d84
imageID: docker-pullable://quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
image: quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
imageID: docker-pullable://quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
image: sha256:10639c807879e82aa694ca1189c8fd06844fcdc7120ecf15cc367021bd392d84
imageID: docker-pullable://quay.io/coreos/etcd-operator@sha256:c0301e4686c3ed4206e370b42de5a3bd2229b9fb4906cf85f3f30650424abec2
Issue is reproduced # openshift version openshift v3.11.0-0.25.0 ose-console image ose-console-v3.11.0-0.25.0.0 Tagged Haseeb - Do you know the status of the downstream etcd operator images? This problem still exist in latest ocp: openshift v3.11.0-0.28.0 I believe I know the source of this issue.
When you install a new operator, OLM installs CRDs for you. In this case, EtcdCluster. The UI knows which resources it can display based on the discovery endpoint in kube. Unfortunately, there's no way to watch the discovery endpoint (like you can actual kube resources) and we have to rely on polling to find the new types.
This problem is solved in the UI by refreshing the page. I'll try to find out what the latest solution for this is in the UI. If it's polling discovery or manual refreshing, I'm wondering if the 404 page should just include language for that ("if you just installed a new operator / CRD / API Type, you may need to refresh")
Alec has a PR that should fix this: https://github.com/openshift/console/pull/497 Note that the PR relies on polling for non-admin users, so this might still be a problem. I wonder if we should perform discovery again if the user navigates to a resource we don't recognize. The CLI does something similar. I tried with non-admin user, and in ui, operator section, got unexpected error: configmaps is forbidden: User "jianzhangbjz" cannot list configmaps in the namespace "operator-lifecycle-manager": no RBAC policy matched (In reply to ge liu from comment #10) > I tried with non-admin user, and in ui, operator section, got unexpected > error: configmaps is forbidden: User "jianzhangbjz" cannot list configmaps > in the namespace "operator-lifecycle-manager": no RBAC policy matched This specific problem is tracked by bug 1625996 The following PR has merged to 4.0, which should make the problem better. It will watch or poll for APIs depending on your RBAC permissions, and in cases where we still don't have the resource, it will show an error message asking the user to refresh. https://github.com/openshift/console/pull/497 The bug:1625996 have been fixed and verified in openshift v3.11.7, and I tried non-admin user: "# oc adm policy add-cluster-role-to-user edit geliu" following the bug:1625996 comment, the error reported in comment 10 disappears, but 404 error still exist when create "etcd clster"/"etcd backup/restore". I also encounter this issue when creating the Couchbase cluster. Workaround: refresh this page *** Bug 1638705 has been marked as a duplicate of this bug. *** (In reply to ge liu from comment #12) > The bug:1625996 have been fixed and verified in openshift v3.11.7, and I > tried non-admin user: "# oc adm policy add-cluster-role-to-user edit geliu" > following the bug:1625996 comment, the error reported in comment 10 > disappears, but 404 error still exist when create "etcd clster"/"etcd > backup/restore". This bug was fixed in 4.0, and it looks like you've tried to verify in 3.11. Can you confirm you tested a 4.0 build? If not, can you try this in 4.0? Verified with 4.0.0-0.nightly-2019-03-06-074438. Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:0758 |