Description of problem:
The discover-etcd-initial-cluster tool should archive the data-dir if the target member is unstarted. This is useful for disaster recovery after restoring a backup.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Bring up a single node cluster by restoring the backup on a single master node using the restore DR script.
2. Force CEO redeployment via oc patch forceredpeloymment=reason-time
3. When CEO redeployment is forced, the existing nodes will start with new pods, a scale up similar to bootstrap will take place
4. Actual results:
The other two nodes stay Unstarted because of the existing /var/lib/etcd/member
The scale up occurs correctly.
Verified on 4.5 0306, the etcd data dir: /var/lib/etcd, could not exist after etcd restore, otherwise, the etcd member pods will not be scale up, only etcd-member.yaml file is necessary.
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.