[Halld-jlab] Group disk full

[Mark M. Ito]

David et al.,

I just deleted another release of sim-recon from the disk. Now we have 
the most recent three stored there.

Generally, we do not want to store "data" on the group disk, since often 
the volume of output of some data acquisition system is open-ended. The 
usual solution is to write to the work disk and archive to tape.

I posted the current directory usage of the group disk, sorted by size 
at https://halldweb1.jlab.org/docs/group_disk_usage.txt .

   -- Mark

On 12/02/2010 05:29 PM, David Lawrence wrote:
>
> Hi Mark and Yi,
>
>     Beni just told me that the halld group disk is nearly full. I am 
> trying to identify where the space is being used, but it is extremely 
> slow going. I can see one 2GB file in the Coil 1 Cooldown logs 
> directory that looks like it may have been put in today. This is not 
> necessarily to blame, but it may have pushed us up to the 45GB disk 
> limit. Since the coil tests are currently the most active project, I'm 
> wondering if this, and perhaps other things are getting stored on the 
> group disk, contributing to the issue. If so, then we need to find a 
> solution so as to not disrupt the coil test nor other operations that 
> require the group disk.
>
>     Any advice out there?
>
> Regards,
> -David