OpenStack-Ansible

Pike update causes parallel restart of Galera containers

Bug #1736726 reported by Adrien Cunin
10
This bug affects 2 people
Affects Status Importance Assigned to Milestone
OpenStack-Ansible
Fix Released
High
Jesse Pretorius

Bug Description

This backport https://git.openstack.org/cgit/openstack/openstack-ansible-lxc_container_create/commit/?h=stable/pike&id=05b84528d100ab73b91e119dd379a8ed0726db7d contain changes in the tasks "Write default container config" and "Add bind mount configuration to container" that trigger containers restart, including Galera containers restart. If you have multiple Galera containers, it will restart all of them in parallel, thus breaking the Galera cluster.

Had this issue using OSA stable/pike, updating from 27fa064a0eb2c2b17c361e176abe06375f8dc55f (16.0.4) to 16c69046bfd90d1b984de43bc6267fece6b75f1c (head as of two days ago).

Jean-Philippe Evrard (jean-philippe-evrard)
Changed in openstack-ansible:
status: New → Confirmed
importance: Undecided → High
assignee: nobody → Jesse Pretorius (jesse-pretorius)
Revision history for this message
Dmitriy Rabotyagov (noonedeadpunk) wrote :

This has been fixed in later releases for sure. We can't backport fix to pike though as of today as it's EOL now.

Changed in openstack-ansible:
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.
Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.