Remove ClusterVersion channel #1254
Conversation
CI and nightly releases are not part of the official Red Hat Cincinnati graphs. This commit removes the channel property [1], which will result in the NoChannel condition [2], but will keep the CVO from attempting to find its current version in the official Cincinnati [3]. That in turn should keep the CVO from throwing a new, critical alert [4], which will keep it from running afoul of recent update e2e logic that forbids critical alerts [5]. See this related PR [6] which does this for other platforms IPI install step. [1]: https://github.com/openshift/installer/blob/4eca2efd615f8abd65f576721e2410b19f0d40d0/data/data/manifests/bootkube/cvo-overrides.yaml.template#L8 [2]: https://github.com/openshift/cluster-version-operator/blob/fa452c2d270f1f989f3868ef97ae8cf825713583/docs/user/status.md#nochannel [3]: https://bugzilla.redhat.com/show_bug.cgi?id=1827378#c4 [4]: openshift/cluster-version-operator#357 [5]: openshift/origin#24786 [6]: openshift/release#8631 Co-authored-by: W. Trevor King <[email protected]>
|
@wking: changing LGTM is restricted to collaborators DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
/retest |
|
I wonder if we're starting to hit disk space issues? I'm not entirely sure what this error means: /test e2e-metal-ipi |
Similar issues reported for the -upgrade job via https://bugzilla.redhat.com/show_bug.cgi?id=1968754 - I was wondering if it was specific to upgrade but it sounds like probably not? We could increase the disk size ref #1257 but it'd be good to understand why that's become required now |
|
/retest |
|
|
||
| # For CI and nightly releases we need to remove the channel property to | ||
| # prevent critical alerts. | ||
| sed -i '/^ channel:/d' "${assets_dir}/manifests/cvo-overrides.yaml" |
There was a problem hiding this comment.
Should this be conditional, since we do support a release type of "ga" in addition to the "ci" and "nightly" options?
https://github.com/openshift-metal3/dev-scripts/blob/master/common.sh#L117
|
/test e2e-metal-ipi |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: hardys The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-ci
bot
added
the
approved
|
/lgtm |
6 participants
CI and nightly releases are not part of the official Red Hat
Cincinnati graphs. This commit removes the channel property 1,
which will result in the NoChannel condition 2, but will keep the
CVO from attempting to find its current version in the official
Cincinnati 3. That in turn should keep the CVO from throwing a new,
critical alert 4, which will keep it from running afoul of recent
update e2e logic that forbids critical alerts 5.
See this related PR 6 which does this for other platforms IPI
install step.
Co-authored-by: W. Trevor King [email protected]