OpenStack Masakari Charm

Resource: res_masakari_haproxy not running

Bug #1926445 reported by Narinder Gupta on 2021-04-28

This bug report is a duplicate of: Bug #1874719: [SRU] Use the hostname as the node name instead of hardcoded 'node1'. Edit Remove

This bug affects 2 people

Affects		Status	Importance	Assigned to	Milestone
	OpenStack Masakari Charm	Incomplete	Undecided	Unassigned

Bug Description

After implementing the HA using corosync and pacemaker we found the status of masakri is in blocked state. juju status Resource: res_masakari_haproxy not running
It seems node1 was added as part of the HA process and after removing the node1 extra entries from crm everything gets cleared and status show ok.

juju run --unit masakari/0 'sudo crm node delete node1'

Revision history for this message

Alex Kavanagh (ajkavanagh) wrote on 2021-04-30:

Hi Narinda. Please could you provide a few more details about the environment that this was running on:

e.g.
Ubuntu version
charms version
juju version
bundle (sanitised) that reproduces the issue.

Many thanks!

Changed in charm-masakari:
status:	New → Incomplete

Revision history for this message

Narinder Gupta (narindergupta) wrote on 2021-06-24:

seems like duplicate of this bug https://bugs.launchpad.net/charm-hacluster/+bug/1874719

Revision history for this message

Hybrid512 (walid-moghrabi) wrote on 2021-07-05:

I also encounter that status from time to time.
Most of the time, rebooting the unit fixes the issue (juju ssh masakari/<unit> sudo reboot).
I don't know why it does this though and this is not the issue I encounter most (I have exactly the same kind of error which happens a lot more often about Resource: res_masakari_4cc6744_vip not running).

Ubuntu version : 20.04.2 "Focal"
Charms version :
- Masakari v9.0.0 rev #10
- hacluster rev #76
Juju version : 2.9.5

Revision history for this message

Alex Kavanagh (ajkavanagh) wrote on 2021-07-06:

> seems like duplicate of this bug https://bugs.launchpad.net/charm-hacluster/+bug/1874719

Yes, I'm pretty sure that's the case. Going to mark this as a duplicate.

@hybrid512, I think your issue might be different, unless you also have the node1 issue?