Fix invalid olm.skipRange semver in percona-postgresql-operator#9684
Fix invalid olm.skipRange semver in percona-postgresql-operator#9684joelanford wants to merge 1 commit into
Conversation
Pipeline SummaryPipeline: operator-hosted-pipeline Tasks
Pipeline logs: https://gist.github.com/rh-operator-bundle-bot/796d31d93453b077b5f0bca105a8b297 TroubleshootingPlease refer to the troubleshooting guide. Run |
Remove 'v' prefix from version in olm.skipRange annotations. Semver ranges use '<X.Y.Z' not '<vX.Y.Z'. Signed-off-by: Joe Lanford <joe.lanford@gmail.com>
28103f6 to
3fe63c3
Compare
Pipeline SummaryPipeline: operator-hosted-pipeline Tasks
Pipeline logs: https://gist.github.com/rh-operator-bundle-bot/57d40042245532cb45d47eb890b19d4f TroubleshootingPlease refer to the troubleshooting guide. Run |
|
@joelanford Hello. I noticed that you tried to modify existing bundles, same as in this PRs -> #9686 #9687 It is not possible to edit existing bundles unless you would switch/onboard to FBC - onboarding doc |
|
I'm from the OLM team, not a persona developer, so I can't make that decision. I audited all OLM catalogs and noticed this problem with lots of bundles. We need to update opm validate to reject invalid skip ranges, but we can't do that until all existing issues are fixed. Otherwise, this catalog will permafail on opm validate when the pipeline updates to use it. Anything we can do to fix this without forcing the persona folks to adopt FBC quickly? |
|
@joelanford I am afraid that without switching to FBC there is no easy way how to modify existing, released bundles. Since the needed change is part of the CSV metadata and it requires rebuilding the bundle image and releasing it, it is not possible to do it unless they would onboard to FBC. |
|
This pull request is stale because it has been open for 30 days with no activity. |
Summary
vprefix from version inolm.skipRangeannotations across 12 CSV versions<vX.Y.Zis not valid semver range syntax; the correct form is<X.Y.Z