Hi, I ran into a fleet related issue and I hope so...
# fleet
m
Hi, I ran into a fleet related issue and I hope somebody could give me a hint on how to fix this. I’m running a rancher multicluster setup and usually the fleet-agent gets deployed to the remote cluster automatically whenever I import a new cluster to rancher. Due to another weird issue of fleet on the remote clusters I tried to remove the remote cluster from rancher and import it again. Prior to connecting the remote cluster again I tried to clean up all fleet Resources in the rancher cluster that I could find to be associated with the specific remote cluster. I also deleted the fleet-agent-<clustername> Bundle and BundleDeployments, Clusterregistrations, … Resources. My problem now is that when I import this remote cluster again with the same name, the cattle-cluster-agent on the remote cluster connects to rancher, creates the corresponding cluster.fleet.cattle.io Resource, etc. but on the actual remote cluster the fleet-agent does not get installed anymore and not able to deploy out of rancher’s continuous deployment section. So it seems that I missed some data during my cleanup and therefore fleet has some dangling data about the original remote cluster that prevents a new deployment of the fleet-agent to the remote cluster. Unfortunately, after many tries, I couldn’t find any more Resource that I may have forgotten to cleanup additionally. I probably could use a new name for the imported cluster, but we already have a significant amount of GitRepos and ClusterGroups targeting the remote cluster by its original name, so I think it would be much better if I could fix this instead of working around the issue.