Rook: Monitors still don't use letter consecutively

Created on 5 Oct 2020  路  4Comments  路  Source: rook/rook

Is this a bug report or feature request?

  • Bug Report

Deviation from expected behavior:

I think some time ago it was implemented that monitors were chosen letter consecutive indexes. While it looks like it does not really work.

This morning I had a little outage and out of 3 monitors: mon-a, mon-b, mon-e - the latter mon-e was unavailable. So after 10 minutes deadline mon-f was created. So I now run in a cluster with a, b, f.

What happens when I run out of alphabet?

Expected behavior:
I think letters should be assigned consecutively, if possible. In the case above it should be possible.

How to reproduce it (minimal and precise):

Make a monitor unavailable several times, and wait while it fails-over.

File(s) to submit:

  • Cluster CR (custom resource), typically called cluster.yaml, if necessary
  • Operator's logs, if necessary
  • Crashing pod(s) logs, if necessary

    To get logs, use kubectl -n <namespace> logs <pod name>
    When pasting logs, always surround them with backticks or use the insert code button from the Github UI.
    Read Github documentation if you need help.

Environment:

  • OS (e.g. from /etc/os-release):
  • Kernel (e.g. uname -a):
  • Cloud provider or hardware configuration:
  • Rook version (use rook version inside of a Rook Pod): rook: v1.3.5
  • Storage backend version (e.g. for ceph do ceph -v): ceph version 14.2.9 (581f22da52345dba46ee232b73b990f06029a2a0) nautilus (stable)
  • Kubernetes version (use kubectl version):
  • Kubernetes cluster type (e.g. Tectonic, GKE, OpenShift):
  • Storage backend status (e.g. for Ceph use ceph health in the Rook Ceph toolbox):
bug

All 4 comments

@zerkms When the single-letter alphabet is exhausted, a double-letter name will be used, then triple-letter, etc. The only time the letters increase is when there is a mon failover. What do you see is not working? Rook can't use the same mon name after a failover.

@travisn I think I seen it in some other issue report several years ago: that in the deployment a, b, e, on failover c would be used (to replace dead e), as it's available now. Or am I confusing it with something else?

@zerkms Rook has always added the next mon in alphabetical increasing order. There was a bug a few months back that caused rook to get confused about the index of the mon last used, but that was fixed.

Okay, so if exhausting is not causing any troubles - I don't mind. Thanks! :-)

Was this page helpful?
0 / 5 - 0 ratings