Damien Ciabrini 9602a9bafc Instance HA: prevent compute to start on a host being evacuated ... When the evacuation of a nova-compute is in progress on a compute node, starting the nova-compute service on that node would cause a race in setting task_state [1] and could ultimately lead failures with all future evacuation attempts on the compute node. To avoid triggering that situation, introduce a startup wrapper when Instance HA is enabled. The wrapper loops until the force-down flag is clear and all evacuation finished, and then starts nova_compute. [1] If71727cde51c29231dbb9a51c5babbcdfc802bdd Partial-Bug: #1764883 Co-Authored-By: Andrew Beekhof <abeekhof@redhat.com> Change-Id: Id1fc820b42fb72fc861fda82b04f6a3fa2b6b6f6		2018-05-03 07:05:39 +02:00
..
instanceha	Instance HA: prevent compute to start on a host being evacuated	2018-05-03 07:05:39 +02:00
ssh	Change template names to queens	2017-11-23 10:15:32 +01:00
pacemaker_common_functions.sh	Maintain ceph-osd package only on nodes hosting CephOSD service	2017-09-05 22:55:29 +00:00
pacemaker_maintenance_mode.sh	Restart haproxy after configuring SSL certs	2016-04-01 12:42:02 -04:00
pacemaker_resource_restart.sh	Fix external Load Balancer deployment	2016-11-15 11:25:38 +01:00
post_puppet_pacemaker.j2.yaml	Change template names to queens	2017-11-23 10:15:32 +01:00
post_puppet_pacemaker_restart.yaml	Change template names to queens	2017-11-23 10:15:32 +01:00
pre_puppet_pacemaker.yaml	Change template names to queens	2017-11-23 10:15:32 +01:00
run_puppet.sh	puppet apply: add --summarize	2018-01-04 09:37:46 -08:00