Closed alban closed 6 years ago
E0612 12:16:54.377420 1 leaderelection.go:224] error retrieving resource lock kube-system/etcd-operator: Get https://10.241.0.1:443/api/v1/namespaces/kube-system/endpoints/etcd-operator: dial tcp 10.241.0.1:443: getsockopt: connection refused
10.241.0.1
is the service address of the Kubernetes API. There is an unknown issue with failed apiserver pods that I'm currently looking into (hence connection refused
). It can lead to different errors during setup, the error above is just one of the symptoms that we observe.
./wait_for_cluster.sh: line 103: fail: command not found
I created a separate issue for that: https://github.com/gardener/landscape-setup/issues/11
Is this issue related to kubify or the landscape setup project?
@Diaphteiros can you comment on that one?
@afritzler as far as I have seen, that was also due to https://github.com/gardener/kubify/issues/49 and can be closed as well.
Understood. Well, I would suggest closing this one here and opening an issue in the landscape setup.
@afritzler That issue is related to kubify, but I also think it could have been connected to https://github.com/gardener/kubify/issues/49 and be resolved by now. I should probably update the kubify submodule in the landscape setup project soon to see whether this problem still occurs.
Closing this one since the this should be resolved with #49.
I followed the instructions in landscape-setup-template and
deploy_kubify.sh
fails to deploy etcd:More logs:
I got more logs from the master 0:
The process etcd-operator is running but no further etcd processes. The Docker logs of the etcd-operator container are the following:
And the errors go on for a while.
On master 1, neither etcd-operator or etcd are running.
On master 2, etcd is running but cannot complete an election because it's alone.