Versions Affected
3.6.0 and below
Summary
New jobs were all remaining in a "pending" state
Ssh'd to the CB server and check for a job lock file
[root@cloud cloudbolt]# ls -al /opt/cloudbolt/runjobs.lock
-rw-r--r-- 1 root apache 0 Feb 20 13:47 /opt/cloudbolt/runjobs.lock
Note that the lock file is 6 days old, and that no "run jobs" process is currently running
Check the update on the CloudBolt server
[root@cloud cloudbolt]# uptime
12:42:09 up 1 day, 1:28, 1 user, load average: 0.67, 0.63, 0.42
This implies that the CloudBolt VM was rebooted while the job engine was running some jobs, and the lock file never got cleaned up.
Workaround
Do not reboot the CloudBolt Server VM when there are jobs running.
Fix
A future version of CloudBolt will detect that a lockfile needs to be cleaned up after a restart.
0 Comments