This message was deleted.
# fleet
a
This message was deleted.
n
Fixed afte changing
fleet-default
to
fleet-local
1
c
Have you deployed Rancher Manager in a docker container or on a k8s cluster (K3s for ex.) ?
n
Rancher has been installed using these steps: https://github.com/halkyonio/rancher#install-rancher
c
ok K3s
n
I'm a bit lost with all the acronyms: RKE, k3s, RKE1,2,3
c
the short story: RKE: legacy kubernetes engine, docker based RKE2 new kubernetes engine architecture (inspiured by K3s architecture) and containerd based K3s lightweight Kubernetes engine
if you start, stay away from rke, choose between K3s and RKE2 whether you prefer 'lightweight' or 'full' K8s
n
So, when I execute
curl -sfL <https://get.k3s.io> | sh -s - server
I install a k3s server (+ tools + service) and k3s agent (= what is used to create on different providers additonal k8s clusters) - correct ?
c
k3s server is for the control plane of the cluster and k3s agent are 'workload only' nodes
g
i think you need to specify a cluster selector.. the default behaviour is nil.. https://fleet.rancher.io/gitrepo-structure/#per-cluster-customization
n
Thanks
t
We're also having the same problem. The repos and clusters are in the same NS, and I've added clusterSelector: {} , but each repo in Fleet is showing "There are no clusters available"
g
what does the bundle look like? also what is response of
kubectl get clusters.fleet -A
t
Many apologies! I did eventually fix the problem. We are moving to a new install of Rancher and did have the git credentials/URIs exactly the same. Once I fixed that up, it started to work. Thanks for the reply, though!
430 Views