[Continous Delivery] GitRepos report that state has changed on all clusters when just one has changed

[it-slav]

What kind of request is this (question/bug/enhancement/feature request): Bug

Steps to reproduce (least amount of steps as possible): 3 clusters managed by Fleet. An app is deployed to all 3 At one of the clusters the replicas is changed.

Result: The gui shows that all 3 clusters has a modified state, see screenshoot. Here I have locally modified cluster rkes replicas for frontend deployment.

Other details that may be helpful: Expected that gui show that only one cluster (named rke) has modified state.

Environment information

• Rancher version (rancher/rancher/rancher/server image tag or shown bottom left in the UI): 2.5.5
• Installation option (single install/HA):Single

[gregkeys]

I have a similar situation but everything in continuous delivery shows modified and wont reach a non modified state.

[jiaqiluo]

I hit a similar issue on v2.5.8-rc9 with fleet version: v0.3.5-rc6

Steps:

• run rancher single-install and provision 3 single-node clusters
• install the example multi-cluster Helm app into all 3 clusters

Results: The resources page show all 3 services are pending, but in fact, only 1 of them is pending because it is waiting for LB to be ready. I can confirm the other 2 servers are active in the target clusters.

[kkaempf]

Seems to be an UI issue, dropping team/fleet

[rak-phillip]

@gaktive transferring to rancher/dashboard for triage