[BUG] This git repo is not targeting any clusters

[next0262]

Is there an existing issue for this?

• I have searched the existing issues

Current Behavior

After upgrading rancher from v2.7.6 to v2.7.7(fleet v0.7.1 -> v0.8.0), the created gitrepo is not able to find the clusters matched with clusterGroup.

It just shows "This git repo is not targeting any clusters"

Expected Behavior

GitRepo should show the clusters of the clusterGroup and deploy the charts & manifests of the github repo

Steps To Reproduce

1. Upgrade rancher from v2.7.6 to v2.7.7
2. Create GitRepo with github repo url and target clusterGroup
3. GitRepo shows "This git repo is not targeting any clusters"
   BUNDLEDEPLOYMENTS-READY: 0/0

Environment

- Architecture: x86_64
- Rancher Version: v2.7.7
- Fleet Version: v0.8.0
- Cluster:
  - Provider: RKE2
  - Options: 
  - Kubernetes Version: v1.26.6+rke2r1

Logs

No response

Anything else?

No response

[lindhe]

I think I have the same issue. But not quite sure, perhaps it's just really slow (like 1 hour for updating status); it seems like it resolves itself eventually. I think. Certainly buggy behavior, though.

[nitin992vij]

got the same issue "This git repo is not targeting any clusters" with the clusterGroup & Cluster with Rancher Version: 2.7.6

[lindhe]

Turns out we had some network issues yesterday. I think that what happened in my case is that the pull operation took a really long time, and until that was finished Fleet reported "This git repo is not targeting any clusters", presumably because it cannot create any bundles before it has successfully pulled the resources. Does this make sense, considering how Fleet works internally?

If so, I think the status message should change to something more descriptive, e.g. "Fetching bundle resources" or at least "The bundle is waiting to become ready".

[kkaempf]

@raulcabello jfyi as you're working on gitjob these days ...

[dalbertom]

We are having this issue as well. Shortly before we noticed the issue we had just upgraded to Rancher v2.10.0 (Fleet v0.11.1), but after rolling back to Rancher v2.9.2 (Fleet v0.10.2) the issue persisted. We were also exploring the option of using webhooks rather than polling for the gitrepos, but that was rolled back as well.

The symptoms we see are very similar to the ones described in this ticket, the GitRepo shows "This git repo is not targeting any clusters" and $ kubectl -n fleet-default get gitrepos shows BUNDLEDEPLOYMENTS-READY 0/0

Additionally, a lot of our bundles are stuck in "InProgress", "Removing" or "OutOfSync" state but when opening them to see their resources they are all in "Ready" (for InProgress and OutOfSync) or no resources at all in the case of Removing.

If there are any logs needed to provide more details I'd be happy to send those along.

Thanks

[und-miller]

Also experiencing this issue