Keystone token flush running simultaneously on all containers
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack-Ansible |
Fix Released
|
Medium
|
Andy McCrae |
Bug Description
It seems that the token flush job is running on all keystone containers at the same time, unlike it used to be in older versions, where it was installed only on the first container :
# ansible keystone_container -m shell -a 'cat /var/spool/
# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (/tmp/crontabTWwI1U installed on Tue Sep 27 15:56:15 2016)
# (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
#Ansible: Clear out stale keystone tokens
0 * * * * /openstack/
infra03_
# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (/tmp/crontabCaRbz2 installed on Tue Sep 27 15:56:15 2016)
# (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
#Ansible: Clear out stale keystone tokens
0 * * * * /openstack/
infra01_
# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (/tmp/crontab0qdmRZ installed on Tue Sep 27 15:56:15 2016)
# (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
#Ansible: Clear out stale keystone tokens
0 * * * * /openstack/
Why was this changed ? To overcome HA issues in case on container is not operational ?
Changed in openstack-ansible: | |
importance: | Undecided → Medium |
Fix proposed to branch: master /review. openstack. org/433744
Review: https:/