Re: [Condor-users] Using DAGManJobID for job priority

["Peter F. Couvares" <pfc@xxxxxxxxxxx>]

This is now fixed for 6.8.1 -- specifically, the bug preventing $ 
(DAGManJobId) from being expanded in DAG node submit files has been  
addressed.  Thanks to Michael for the report.

You can now use DAGManJobId as the value of your job's queue  
priority, like so:

 Priority = $(DAGManJobId)

This will ensure that jobs submitted by earlier DAGMans will have  
local, per-user queue priority over jobs submitted by later DAGMans.

-Peter


On Jul 26, 2006, at 11:22 PM, Peter F. Couvares wrote:
> I'm looking into this.  It may be a bug for 6.8.1...
>
> Thanks,
>
> -Peter
>
>
> On Jul 26, 2006, at 5:59 PM, Michael S. Root wrote:
>
> > Hi, I've just upgraded to 6.8.0, and I'm trying to figure out how  
> > to use
> > the DAGManJobID attribute to set the job priority of dag children.
> >
> > I actually mailed the list a while ago about exactly what I'm  
> > trying to
> > do:
> > https://lists.cs.wisc.edu/archive/condor-users/2005-February/ 
> > msg00089.shtml
> >
> > I see this in the release notes:
> >
> > # The DAGManJobID attribute which condor_dagman inserts into the  
> > classad
> > of every job it submits now contains only its cluster ID (instead  
> > of a
> > cluster.proc ID pair), so that it may be referenced as an integer  
> > in DAG
> > job submit files. This allows, for example, a user to  
> > automatically set
> > the relative local queue priority of jobs based on the  
> > condor_dagman job
> > that submitted them, so that jobs submitted by ``older'' DAGs will  
> > start
> > before jobs submitted by ``newer'' DAGs (assuming they are otherwise
> > identical).
> >
> > This seems to be exactly what I want, but I can't get it to work.   
> > If I add:
> >
> > priority = DAGManJobID
> >
> > or
> >
> > priority = "DAGManJobID"
> >
> > to a sub-job submit file, the classAd for the job when it's  
> > running just
> > says that JobPrio=0.
> >
> > Can anyone give me an example?  I feel like I'm missing something  
> > obvious.
> >
> > -Mike