Re: Timeout setting for ClientRunBeforeJob

Bill Arlofski via Bacula-users <[email protected]> Wed, 18 Jun 2025 14:41:57 +0000

Newsgroups	gmane.comp.bacula.user
Message-ID	<[email protected]>

On 6/18/25 5:59 AM, Yateen Shaligram Bhagat \(Nokia\) via Bacula-users wrote:
> My bad, I made a wrong statement in my earlier email.
> 
> Here is the corrected scenario.
> 
> I have a dynamic fileset creation with a script that runs on the client (Linux)  side with File resource, see the setting below:
> 
>                   File = "\\| su - username -c “/path/to/the/script”
> 
> The execution of this script takes a long time, resulting in timeout , thus failing the backup job
> 
> *bacula-server-dir JobId 1101693: Fatal error: Bad response to Storage command: wanted 2000 OK storage*
> 
> *, got 2902 Bad storage |*
> 
> *FD termination status:  Error*
> 
> *                 SD termination status:  Waiting on FD*
> 
> Thanks
> 
> Yateen

Hello Yateen,

This is a known issue.

It has only recently started becoming an issue because we are seeing people doing this and, for example, are using the `find` 
command in a dynamic FileSet to traverse larger and larger filesystems in HPC environments.

Due to exploding data growth, their scripts (like yours), or their find commands in their dynamic FileSets are starting to 
take longer and longer and this issue has surfaced.

If I remember correctly, what is happening is that the SD times out waiting for the FD, and gives up. By the time the FD 
starts to send data, the job is not running on the SD, hence the `Bad storage` response.

I am pretty sure some work to mitigate this is already in progress - at least I know it has been discussed recently. :)

Best regards,
Bill

-- 
Bill Arlofski
[email protected]

_______________________________________________
Bacula-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bacula-users

signature.asc (application/pgp-signature, 583 B)