[UBUNTU 18.04.4] Charmed Kubernetes local deployment failed for s390x because of etcd error

[bugproxy]

Steps to reproduce:

Clean Z LPAR machine. Followed official documentation https://ubuntu.com/kubernetes/docs/install-local.

Actual results:
Cluster did not start and hang in the follow state:
# juju status
Model Controller Cloud/Region Version SLA Timestamp
k8s lxd-controller localhost/localhost 2.8.1 unsupported 16:11:10+02:00

App Version Status Scale Charm Store Rev OS Notes
containerd 1.3.3 maintenance 5 containerd jujucharms 80 ubuntu
easyrsa 3.0.1 active 1 easyrsa jujucharms 318 ubuntu
etcd 3.3.15 waiting 3 etcd jujucharms 521 ubuntu
flannel 0.11.0 blocked 5 flannel jujucharms 492 ubuntu
kubeapi-load-balancer 1.14.0 active 1 kubeapi-load-balancer jujucharms 730 ubuntu exposed
kubernetes-master 1.18.5 waiting 2 kubernetes-master jujucharms 850 ubuntu
kubernetes-worker 1.18.5 blocked 3 kubernetes-worker jujucharms 682 ubuntu exposed

Unit Workload Agent Machine Public address Ports Message
easyrsa/0* active idle 0 10.63.81.207 Certificate Authority connected.
etcd/0* active idle 1 10.63.81.209 2379/tcp Errored with 0 known peers
etcd/1 waiting idle 2 10.63.81.36 Waiting to retry etcd registration
etcd/2 waiting idle 3 10.63.81.218 Waiting to retry etcd registration
kubeapi-load-balancer/0* active idle 4 10.63.81.145 NGINX is ready
kubernetes-master/0* waiting executing 5 10.63.81.239 Waiting for kube-api-endpoint relation
  containerd/4 maintenance executing 10.63.81.239 (install) installing charm software
  flannel/4 blocked executing 10.63.81.239 (install) Waiting for etcd relation.
kubernetes-master/1 waiting executing 6 10.63.81.213 Waiting for kube-api-endpoint relation
  containerd/3 active executing 10.63.81.213 (leader-settings-changed) Container runtime available
  flannel/3 blocked executing 10.63.81.213 (leader-settings-changed) Waiting for etcd relation.
kubernetes-worker/0 waiting executing 7 10.63.81.155 Waiting for cluster DNS.
  containerd/2 active executing 10.63.81.155 (start) Container runtime available
  flannel/2 maintenance executing 10.63.81.155 Negotiating flannel network subnet.
kubernetes-worker/1 blocked executing 8 10.63.81.53 (config-changed) Connect a container runtime.
  containerd/1 active executing 10.63.81.53 (start) Container runtime available
  flannel/1 maintenance executing 10.63.81.53 Negotiating flannel network subnet.
kubernetes-worker/2* blocked executing 9 10.63.81.219 Connect a container runtime.
  containerd/0* active idle 10.63.81.219 Container runtime available
  flannel/0* maintenance executing 10.63.81.219 Negotiating flannel network subnet.

Machine State DNS Inst id Series AZ Message
0 started 10.63.81.207 juju-6e9ea7-0 bionic Running
1 started 10.63.81.209 juju-6e9ea7-1 bionic Running
2 started 10.63.81.36 juju-6e9ea7-2 bionic Running
3 started 10.63.81.218 juju-6e9ea7-3 bionic Running
4 started 10.63.81.145 juju-6e9ea7-4 bionic Running
5 started 10.63.81.239 juju-6e9ea7-5 bionic Running
6 started 10.63.81.213 juju-6e9ea7-6 bionic Running
7 started 10.63.81.155 juju-6e9ea7-7 bionic Running
8 started 10.63.81.53 juju-6e9ea7-8 bionic Running
9 started 10.63.81.219 juju-6e9ea7-9 bionic Running

After ssh to etcd/1 I can see error logs at /var/log/syslog:
Jul 14 14:11:46 juju-6e9ea7-2 etcd.etcd[27280]: 2020-07-14 14:11:46.612925 E | etcdmain: etcd on unsupported platform without ETCD_UNSUPPORTED_ARCH=s390x set.
Jul 14 14:11:46 juju-6e9ea7-2 systemd[1]: snap.etcd.etcd.service: Main process exited, code=exited, status=1/FAILURE
Jul 14 14:11:46 juju-6e9ea7-2 systemd[1]: snap.etcd.etcd.service: Failed with result 'exit-code'.

[George Kraft (cynerva)]

We're aware of this issue and committed a fix back in March, but it looks like we never released it.

See https://bugs.launchpad.net/etcd-snaps/+bug/1869232

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2020-07-16 04:43 EDT-------
Is there a planned target date available, when this problem will be fixed?
Many thx in advance

[George Kraft (cynerva)]

We don't currently have a target date. My boss is out this week, but when he gets in on Monday, I will bring it to his attention. I'm hopeful we can get a release out soon, maybe sometime next week, but no promises.

[Tim Van Steenburgh (tvansteenburgh)]

Fix is in these snaps:

  3.4/stable: 3.4.5 2020-04-30 (230) 22MB -
  3.4/candidate: 3.4.5 2020-04-30 (230) 22MB -
  3.4/beta: 3.4.5 2020-04-30 (230) 22MB -
  3.4/edge: 3.4.5 2020-04-30 (230) 22MB -
  3.3/stable: 3.3.19 2020-07-21 (229) 21MB -
  3.3/candidate: 3.3.19 2020-07-21 (229) 21MB -
  3.3/beta: 3.3.19 2020-07-21 (229) 21MB -
  3.3/edge: 3.3.19 2020-04-07 (229) 21MB -

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2020-08-26 09:09 EDT-------
@Canonical. any update available here?

[Frank Heimes (fheimes)]

This ticket is already marked as Fix Released and therefore closed.
With that a fix was provided and updates were rolled-out already.
See also comment #4.

[bugproxy (bugproxy)]

------- Comment From <email address hidden> 2020-09-04 03:02 EDT-------
IBM Bugzilla closed, Fix Released by Canonical