Re: Filesets: splitting a huge directory

Hello Stefan,


On 1/6/26 10:22, Stefan G. Weichinger via Bacula-users wrote:
> 
> greetings, happy new year to everyone!
> 
> I have a FAQ, I assume, but my googling and RTFM didn't help me so far, sorry.
> 
> At a customer I have to set up various Jobs to backup a huge directory with 
> movies. I want to split that into several smaller jobs by using Filesets.
> 
> So I asked an LLM:
> 
> # this should backup only files and directories
> # starting with "a-g" or "A-G"

Not sure it's the best result you can get :-)

I would probably suggest to just call "ls" from a dynamic fileset command
something such as

Include  {
   File = "\\|sh -c 'ls -d /home/[a-gA-G]*'"
}

To be tested, and if the escaping is causing an issue, it can be stored
in a local script.

Hope it helps!

Best Regards,
Eric

> Fileset {
>    Name = "films-A-G"
>    Include {
>      File = "/data/multimedia/films"
>      Options {
>        Regex = ".*/[A-Ga-g][^/]*$"
>      }
>    }
> }
> 
> That doesn't seem to work, and more or less looks wrong to me.
> 
> What's the way to go here?
> 
> Regex? Wildfile? WildDir?
> 
> trial and error is a bit problematic, as these jobs are terabytes in size quickly.
> 
> Do I remember correctly that someone posted some test-script for such cases?
> 
> thanks in advance ...
> 
> 


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