7f156476f9
In this commit:
- completed orch request is soft deleted right away.
- failed orch requests that have not been soft deleted
are soft deleted at the end of the next audit cycle.
- prior to a sync run, if the managed subcloud has any
queued, in-progress, and/or failed requests, its
sync status will be set to out-of-sync.
- a cron job that runs 20 minutes in the first hour
daily (another commit in stx-puppet repo) will purge
all soft deleted orch requests that are 3 days old,
their orch jobs and resources.
Tests:
- Identity and platform config changes with and
without proxy.
- Database inspection after config changes and after
soak.
- Purge of large number of orch requests, orch jobs
and resources.
Story: 2007267
Task: 39044
Change-Id: If5c75b9af3e1e20c010308919cb6be8f11575546
Signed-off-by: Tee Ngo <tee.ngo@windriver.com>
24 lines
568 B
Bash
24 lines
568 B
Bash
#!/bin/bash
|
|
|
|
#
|
|
# Wrapper script to run dcorch-manage db_clean on active controller only
|
|
#
|
|
re='^[0-9]+$'
|
|
|
|
function is_active_pgserver()
|
|
{
|
|
# Determine whether we're running on the same controller as the service.
|
|
/usr/bin/sm-query service postgres | grep -q enabled-active
|
|
return $?
|
|
}
|
|
|
|
if is_active_pgserver
|
|
then
|
|
if [[ $1 =~ $re ]]
|
|
then
|
|
/usr/bin/nice -n 2 /usr/bin/dcorch-manage db_clean $1 >> /var/log/dcorch/dcorch-clean.log
|
|
else
|
|
/usr/bin/nice -n 2 /usr/bin/dcorch-manage db_clean 3 >> /var/log/dcorch/dcorch-clean.log
|
|
fi
|
|
fi
|