End-to-end SCC flake

[smarterclayton]

[INFO] Validating privileged pod exec
Login successful.

You don't have any projects. You can try to create a new project, by running

    $ oc new-project <projectname>


[FAIL] !!!!! Test Failed !!!!
...
W1026 20:36:42.661037   15154 cmd.go:200] -p POD is DEPRECATED and will be removed in a future version. Use exec POD instead.
!!! Error in hack/../test/end-to-end/core.sh:214
    'grep -q "unable to validate against any security context constraint"' exited with status 0 1
Call stack:
    1: hack/../test/end-to-end/core.sh:214 main(...)
Exiting with status 1
Error from server: User "e2e-default-admin" cannot list all buildconfigs in the cluster

from https://ci.openshift.redhat.com/jenkins/job/merge_pull_requests_origin/3781/console

[deads2k]

@pweil- This comes from the the SCC admission control plugin, right?

[pweil-]

yes

[deads2k]

@pweil- This comes from the the SCC admission control plugin, right?

The admission controller works against a cache. We'd need a unified cache and an access review endpoint to fix this problem.

@pweil- you want me to keep this or would you like to take it?

[smarterclayton]

There was no error output from the command.
Running hack/../hack/../test/end-to-end/core.sh:294: executing 'oc exec -n default -tip router-1-8i66o ls' expecting failure and text 'unable to validate against any security context constraint'...
FAILURE after 0.090s: hack/../hack/../test/end-to-end/core.sh:294: executing 'oc exec -n default -tip router-1-8i66o ls' expecting failure and text 'unable to validate against any security context constraint': the output content test failed
There was no output from the command.
Standard error from the command:

[stevekuznetsov]

@smarterclayton I'm not sure about this, but I recently started seeing this again on Jenkins and added a failure cause pointing to this issue -- did anything ever happen to fix this issue or has it simply laid latent since October? Is this issue still relevant to HEAD?

[smarterclayton]

We had a big slowdown in the API server on the rebase which David is fixing - lots of rare flakes suddenly became less rare.

[deads2k]

This was being flagged by a bad indication that was getting triggered during a cleanup step.