Solved

Cloud Storage alert report mount and prunning errors/issues

  • 22 December 2021
  • 4 replies
  • 1168 views

Userlevel 1
Badge +6

Have some alerts that just started showing up in the past couple of days for a particular AWS cloud library

Cloud MountPath(s) without pruner MediaAgent [ {Library[Lib1] MountPath[[store2] s3-backup] DeviceName[Device_94]} ]  (yesterday and today. Same time each day)

and

Failure Reason: Failed to check status of parent directory, media agent cannot access the storage. (started on the 19th and occurs roughly at the same time each day) (Additionally this error first occurred when the cloud storage was associated with a different Media Agent.  )

The library has two mount paths, and the mount error mostly comes from the 2nd path.  Although it once reported the other.

Checked the jobs for the storage policy related to it, and it seems that nothing is missing.

Have looked through the logs and other than the similar alert text, I don’t see what could be causing the alerts. 

There are multiple storage policies using the same Media Agent and the same S3 bucket.  None of the others are having this problem.

Looking for suggestions on what and where to look for.

icon

Best answer by Damian Andre 23 December 2021, 01:06

View original

4 replies

Userlevel 7
Badge +23

Hey David,

It sounds like you may have inconsistent sharing permissions between Media Agents sharing mount paths in the library causing both issues. One solution is to share each Mount Path in this library with same set of MAs and then set a common MA as pruner MA across all mountpaths. On the mount path properties there is an option to choose which Media Agent controls the data pruning. 

The option is “Select MediaAgents for pruning aged data based on the associated copy's datapath”

Overall, check the sharing options on each mount path and see if read and write access is enables for all MA’s

Userlevel 1
Badge +6

Will give that a try.  Have set that in the main library having the issues.

Since the 15th, the alert has only triggered 9 times.  The main problem child has 5 of them, but it has not alerted since the 22nd.

Will see what the future holds.

Userlevel 1
Badge +6

Hmm.  When I set that on the main problem child, I received an alert “No pruner MA Online Alert Type”

Description: Cloud MountPath(s) without pruner MediaAgent

I had also applied that setting to another one that had triggered in the 29th.  No alert after setting that one.

Update:  Actually this alert has been triggering at approximately 8am every morning since the 22nd for the the main problem child.  Was coincidental that it trigger around the same time I applied the setting

Userlevel 7
Badge +23

Hmm.  When I set that on the main problem child, I received an alert “No pruner MA Online Alert Type”

Description: Cloud MountPath(s) without pruner MediaAgent

I had also applied that setting to another one that had triggered in the 29th.  No alert after setting that one.

Update:  Actually this alert has been triggering at approximately 8am every morning since the 22nd for the the main problem child.  Was coincidental that it trigger around the same time I applied the setting

 

Appreciate the update! Glad that setting seems to have resolved the issue. Worthwhile monitoring your library space utilization to ensure its pruning as expected. you can also check out the deduplication engine chart - pending deletes may have reduced suddenly after adjusting that setting if there was a backlog.

Reply