Liberty to Mitaka upgrade takes down rabbit too long
Bug #1632040 reported by
Bjoern
This bug affects 2 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack-Ansible |
Invalid
|
Low
|
Kevin Carter |
Bug Description
It seems that after the hostname changes, the rabbit cluster is down for as long as the playbooks need to execute in between
https:/
and
https:/
and this can easily be 15-30 minutes or more.
To reduce the time we either should move the restart of the rabbit containers (restart-
Otherwise the API is unnecessarily down for 15-30 minutes,
Changed in openstack-ansible: | |
status: | New → In Progress |
importance: | Undecided → Low |
assignee: | nobody → Kevin Carter (kevin-carter) |
To post a comment you must log in.
Cluster status after reboot:
root@infra01- rabbit- mq-container- 9f6838e1: ~# rabbitmqctl cluster_status infra01- rabbit- mq-container- 9f6838e1' ... [{disc, ['rabbit@ infra01- rabbit- mq-container- 9f6838e1' ]}]}, nodes,[ 'rabbit@ infra01- rabbit- mq-container- 9f6838e1' ]}, name,<< "<email address hidden>">>},
Cluster status of node 'rabbit@
[{nodes,
{running_
{cluster_
{partitions,[]}]