What does the "The medium is already reserved by some other job" error mean?

@ZachHeise , we are currently updating the GUI search to utilize our new search engine, though for now you can search on the docs, kb, or ma pages for MUCH better content.

I searched for the code and found this doc:

https://documentation.commvault.com/11.24/expert/11569_auxiliary_copy_troubleshooting.html

MM2013: Restore job fails when media is reserved for other jobs

Symptom

A restore job fails with the following error message:

Error Code 62:348:

 The media is already Reserved by some other Job[s]

Cause

This error message appears for database restore when the primary copy backup is multi-stream and during the Auxiliary Copy operation the Storage Policy copy is set to Combine streams.

To check the configuration for the Auxiliary copy operation:

1. From the CommCell Browser, select Policies | Storage Policies | <Storage Policy Name>.

2. Right-click the Secondary Copy, and then click Properties.

3. Click the Media tab.

   Note down the number of streams entered in the Combine source data streams to [ ] Stream(s) box.

Resolution

To resolve this issue, you must recover the data from the secondary copy (secondary copy on tape) to secondary copy on the disk, and then perform restores from this secondary copy (configured on disk).

Phase 1: Recover job from Auxiliary Copy on tape library to Auxiliary Copy on disk library

Phase 2: Restore of disk library Auxiliary Copy backup database data to the destination server

• View the job to be restored:

  1. From the CommCell Browser, select Policies | Storage Policies | <Storage Policy Name>.

  2. Right-click the Secondary Copy that contains the data to be restored, point to View, and then click Jobs.

  3. Clear Include All Client Computer check box.

  4. Exclude all clients except the client to be restored.

  5. Clear Specify Time Range, and then click OK.

  6. In the Jobs for Storage Policy Copy window, find the specific job and time range to restore from the job ID and Start Time.

  7. Right-click the job, and then click View Media to assure the correct tapes.

• Create a new storage policy copy with the disk library

  1. Right-click the Storage Policy, point to All Tasks, and then click Create New Copy.

  2. In the Copy Name box, type the name for a copy.

  3. From the Library and the MediaAgent list, select disk library and the MediaAgent hosting the disk library.

  4. Click the Copy Policy tab.

  5. Clear All Backups check box.

  6. In the Backups On and After, select the date before the date of the job to restore.

     For example, a full backup to be restore id 08/18/2013, in the Backups On and After set to date 07/01/2013.

  7. Select Specify Source for Auxiliary Copy check box, and then select the restore job's original secondary copy.

  8. Click the Association tab.

  9. Clear all the clients, and then select only the client being restored.

  10. Click OK to create the policy.

  11. Right-click the newly created secondary copy, point to View and then click Jobs.

  12. Clear Specify Time Range, and then click OK.

  13. The jobs that need to be copied appears in the Jobs for Storage Policy Copy window.

  14. Select all the job that are not to be restored.

  15. Right-click the selected jobs and then click Do Not Copy to prevent these jobs from being copied to the disk library

• Run the Auxiliary Copy

  1. Right-click the Storage Policy, point to All Tasks, and then click Run Auxiliary Copy.

  2. Click Select A Copy, and then select the newly created copy.

  3. An Auxiliary Copy job appears in the Job Controller window which initiates the copy of Job from Tape to Disk.

  4. Once the Auxiliary Copy job is complete, right-click the copy, point to View, and then click Jobs.

  5. Clear Specify Time Range, and then click OK.

  6. Select and right-click the job to be restored, click Browse and Restore to initiate the restore the files required from the job list.

     For more information, refer restore procedure in the agent section.

Thanks Mike, I like your detailed answers. But in this case it is a primary copy, not an Aux copy, that we are having this error code in. Anything on that in the documentation?

Anyway, I didn’t want to necessarily start a case with commvault support - assuming the job eventually finishes; it’s crawled up a single % in the past four hours so I know it’s working, just far slower than usual - but still curious as to what might have caused it.

Hi @ZachHeise 

Is this a tape job? 

If so, It sounds like at some point there are two or more jobs competing for the same media which delayed the job. It may not only be an auxcopy but could have also been a restore, index restore or synthetic full operation. If you investigate the media the job is writing to (view contents on the tape), it may show some other jobs which may have been using it. That could provide you with some clues.

Hi Damian - no, this is all HDD-based backups, no tapes.

@ZachHeise 

Does the job ultimately complete without issue, and the only problem you’re seeing is the error in the Delay Reason field?

If so, was this a one time occurrence or do you see this often on other jobs as well?

Hi Matt,

This is the first time I’ve seen this particular error on a job, and the job is ongoing as we speak. It appears to be continuing (although I don’t know if it will be without issue at this point since it’s not done yet) but it’s far, far slower than it’s ever been before. It started at noon on March 30, so almost 2 days ago, and just in the past 24 hours it’s only progressed from 93% to 97% completed.

While this is our largest file server, the average incremental job for this client typically takes about 3-4 hours, not 48.

Judging from y’all replies, it seems like this behavior is unusual for a HDD, rather than tape based, copy, that is Primary, not an Aux?

Hey @ZachHeise  It sounds to me like there may be an issue reserving resources for this job and it may not be running with the full streams - which could explain why it’s running slower.

Check the streams tab, and note the amount of streams and check the properties of the subclient → Advanced → Performance tab. Check the number of readers configured here. 

Your active streams on the tab in the running job should match the number of readers configured. Additionally for the streams listed in the streams tab of the running job, click on each stream and see if doing so shows an error in the window. 

Good idea. The subclient is configured with 6 readers. The job has 5 active, 1 completed streams. It is the completed stream that has the “ The medium is already reserved by some other job[s].
Source: commserve, Process: MediaManager “ message on it. The other 5 active streams have no message.

It was showing that yesterday too. Shouldn’t the job ‘clear’ that completed stream and start a new one automatically, so that we’d be back up to a maximum of 6 streams?

Does the completed stream show if any data was processed/backed up or does it show 0 bytes?

A backup won’t clear/refresh streams the way an Aux copy would. The backup will scan the content at the start of the job, and then divvy up the data among the configured Readers and each reader backs up it’s piece of the content - when all streams have completed their backup, the job is completed.

Please do!  Reply if it happens again, or anything changes.

I’ll mark your reply as the Best Answer, but I can always undo that 😎