Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

v0.23.0 break operators installation #2914

Closed
cedvan opened this issue Jan 11, 2023 · 5 comments
Closed

v0.23.0 break operators installation #2914

cedvan opened this issue Jan 11, 2023 · 5 comments
Labels
kind/bug Categorizes issue or PR as related to a bug.

Comments

@cedvan
Copy link

cedvan commented Jan 11, 2023

Bug Report

Environment

  • operator-lifecycle-manager version: v0.23.0
  • Cluster type : k3s (from truenas scale)
  • Kubernetes version information: v1.25.3+k3s-9afcd6b9-dirty

Problem

Just install a fresh operator lifecycle manager to new truenas scale server.

  • Install Operator SDK CLI
  • Install OLM : operator-sdk olm install (default last, so v0.23.0)
  • Create cert-manager operator : kubectl create -f https://operatorhub.io/install/cert-manager.yaml => create successfull
  • My cert manager operator never up and result command kubectl get csv -n operators stay empty. operators namespace stay as no resources.

Temporary solution

Uninstall v0.23.0 and force install v0.22.0 with --version v0.22.0. Recreate cert-manager operator, and my cert manager up :)

Conclusion

Version v0.23.0 break operator installation

@cedvan cedvan added the kind/bug Categorizes issue or PR as related to a bug. label Jan 11, 2023
@cedvan cedvan changed the title v0.23.0 don't start operator v0.23.0 break operator installation Jan 11, 2023
@cedvan cedvan changed the title v0.23.0 break operator installation v0.23.0 break operators installation Jan 11, 2023
@awgreene
Copy link
Member

awgreene commented Jan 11, 2023

Ah, this is happening because the community catalogSource isn't being ran in restricted mode, which is fixed in this PR. I'll cut a new release once the PR is merged.

@debovema
Copy link

Same here.
Thank you for the fix, which can be applied directly:

kubectl patch catalogsources.operators.coreos.com -n olm operatorhubio-catalog --type merge --patch '{"spec":{"grpcPodConfig":{"securityContextConfig":"restricted"}}}'

@awgreene
Copy link
Member

The v0.23.1 release fixes this issue. I've marked the v0.23.0 release with a warning.

@leochr
Copy link

leochr commented Jan 12, 2023

@awgreene We are trying to understand the root cause of this problem because this is also affecting our CatalogSource. Is patching of the catalog sources a temporary workaround? Would OLM be updated to handle the CatalogSource (without the restricted mode) differently in the future so it doesn't fail? FYI @git4rk @pbaity

@leochr
Copy link

leochr commented Jan 17, 2023

@awgreene We are trying to understand the root cause of this problem because this is also affecting our CatalogSource. Is patching of the catalog sources a temporary workaround? Would OLM be updated to handle the CatalogSource (without the restricted mode) differently in the future so it doesn't fail? FYI @git4rk @pbaity

@awgreene Please see our questions above. We are trying to understand how best to resolve this issue for our catalog. Should we specify the 'restricted' mode in the CatalogSource? Should the catalog image be rebuilt with a later opm version? How can we make it work consistently across environments (now and later)?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
kind/bug Categorizes issue or PR as related to a bug.
Projects
None yet
Development

No branches or pull requests

4 participants