The master team does not build collection information for individual operators or namespaces. It maintains an inspect command inside of must-gather that does heavy lifting for clusteroperators with related objects. The clusterversionoperator must specify resources it wants collected inside of clusteroperator.status.relatedObjects if must-gather inspect is going to collect it. If the CVO is going to continue not having a clusteroperator (seems like a bug, but ok), you can work with Eric Rich to add to the script that CEE maintains in the must-gather repository.
So looking at the OWNERS for cli: https://github.com/openshift/origin/tree/2819895745315df986ccb74016eb0106f49423b8/pkg/oc/cli/admin/mustgather actual script: https://github.com/openshift/must-gather/blob/bb12f96da940bd0388ca94612a5d7b779a60b891/OWNERS#L6-L10 It seems like Master team does actually own this contrary to https://bugzilla.redhat.com/show_bug.cgi?id=1715634#c1 And must-gather fails to gather namespaces not part of ClusterOperator is not a bug in CVO, rather a bug in the gather command, because it decided to only look at status.relatedObjects for cluster operator, where it could have based the gather on `openshift` prefix for names etc..
Individual components are responsible for their own information gathering. The master team started a repo to create an image that could be use for this purpose, created a generic command to cover the 80% case, started CEE on the path to describing collection, and took it through packaging, but we do not own the individual information gathering specific to non-standard components. The inspect command covers use-cases for normal operators that simply need default actions. If you need more than that (and it appears the CVO does since, it does not conform to the standard), you are responsible for adding it, the same way that the networking team is doing so.
https://github.com/openshift/must-gather/pull/95 should resolve this.
#95 targets the master branch, so I'm moving back to ASSIGNED until we have a backport.
https://github.com/openshift/must-gather/pull/96 should be the PR for 4.1
Not MODIFIED until the backport PR lands.
This pull was reverted. Do not violate process again.
https://github.com/openshift/must-gather/pull/97
https://github.com/openshift/must-gather/pull/95
*** Bug 1722966 has been marked as a duplicate of this bug. ***
Fixed in version: 4.2.0-0.nightly-2019-06-25-003324 # oc get pod -n openshift-cluster-version NAME READY STATUS RESTARTS AGE cluster-version-operator-875f79d4-hvfh4 1/1 Running 0 12m # find |grep openshift-cluster-version ... namespaces/openshift-cluster-version/pods/cluster-version-operator-875f79d4-hvfh4/cluster-version-operator/cluster-version-operator/logs/current.log namespaces/openshift-cluster-version/pods/cluster-version-operator-875f79d4-hvfh4/cluster-version-operator/cluster-version-operator/logs/previous.log ...
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-2019:2922