> -----Original Message-----
> From: [email protected] [mailto:owner-slurm-
> [email protected]] On Behalf Of Michael Di Domenico
> Sent: Tuesday, April 19, 2011 12:26 PM
> To: [email protected]
> Subject: Re: [slurm-dev] JobComp scripts
> 
> Thanks.
> 
> We're using 2.2.4, just upgraded over the weekend.
> 
> Right now i'm using the jobcomp script plugin to push the environment
> variables into a far end database.  we then have a dashboard'ish
> webpage that shows the jobs run on the partitions, by whom, how long
> they took, how many nodes, etc. (the webpage is an aggregate of all
> the clusters on our network (since we use slurm everywhere))

You can get the same effect from the sview tool when only using 1 slurmdbd (the 
preferred way of operation for most sites).  

While not a handy webpage it does work cross cluster.  Also using 1 slurmdbd 
will give your users the power to do multi-cluster options or stat information 
anywhere in the enterprise.  All the SLURM user query tools and sbatch work 
cross-cluster.

> 
> the newest instance of slurm (2.2.4) where we added
> accounting/slurmdbd the mysql database is local to the slurm instance,
> ideally, I'd like to just have my webpage query the slurm (via
> perl/DBI) tables and pull out the necessary info.  but if those
> columns change that often perhaps that's not a good idea

Yeah, it probably isn't a good idea, put you can use the perl apis in already 
in slurm would probably give you what you want.  I know others have done this 
exact thing to do web pages which sound very similar to what you have going.

> 
> On Tue, Apr 19, 2011 at 12:56 PM, Auble, Danny <[email protected]> wrote:
> >
> >
> >> -----Original Message-----
> >> From: [email protected] [mailto:owner-slurm-
> >> [email protected]] On Behalf Of Michael Di Domenico
> >> Sent: Tuesday, April 19, 2011 9:38 AM
> >> To: [email protected]
> >> Subject: Re: [slurm-dev] JobComp scripts
> >>
> >> Danny, we are using the accounting/slurmdbd module to push the
> records
> >> into a database (just starting with it), but it was/is not clear to
> me
> >> that this provides a duplication of the jobcomp information.  If
> >> that's the case, then yes, i could change the jobcomp/mysql plugin
> to
> >> jobcomp/script.
> >
> > Yes, the jobcomp is, outside of the fields you mentioned a small
> subset of what the accounting plugin stores.  There is not much sense
> using both to write to the slurmdbd.
> >
> >>
> >> Since we're on the accounting/slurmdbd subject, for a moment, i
> >> noticed in the accounting table, there is/are a field missing.  in
> the
> >> jobcomp/script output you're setting a "BATCH" environment variable
> to
> >> tell whether a program was produced as a batch or single, that
> column
> >> seems to be missing from the database table.  it would also be nice
> if
> >> this table logged the clustername and what nodes were done on a
> >> NODE_FAIL exit status as additional columns.
> >
> > I am not sure which version of SLURM you are using, but all job
> tables in 2.2+ are prefixed with the name of the cluster, hence why it
> isn't a column in the table.  We don't store anything special about the
> nodes in the job table outside of the nodes actually ran on, but you
> could easily figure out which nodes had issues by using sacctmgr to
> list the events that happened on those nodes at the end of the job.
> >
> > As for the batch flag, if a user uses sbatch or salloc there is a
> flag set called track_steps in the database.  If that isn't set the job
> was started directly from srun.  As for telling if it really was a
> batch script or not, the only way to really tell if that was the case
> it isn't fool proof, but if the user does specify the name of the job
> with salloc the name of the job will be "allocation", you could query
> against that to find batch jobs outside of interactive jobs, but if the
> user states a job name that doesn't work.
> >
> > On another note, I am not sure how you are getting this data to your
> other database, but it is probably a good idea to either use the C api,
> perl api, or sacct[mgr] to get the data since that interface is pretty
> static.  If you are looking at the actual database tables they could
> change often and would probably be difficult to keep up to sync.
>  Especially going from 2.1 to 2.2 where the entire database scheme
> changed dramatically.
> >
> > Danny
> >
> >>
> >> having these extra columns would eliminate my need for a script at
> all
> >> as i could sql pull the data into our site-wide tracking databases
> >>
> >> On Tue, Apr 19, 2011 at 11:59 AM, Danny Auble <[email protected]> wrote:
> >> > Hey Michael, is there any reason you don't want to use the
> >> > accounting_storage/slurmdbd plugin for accounting? It provides an
> >> outrageous
> >> > amount more information about each job and step than the basic
> >> jobcomp
> >> > plugin. It also allows the ability for fairshare scheduling and
> >> limits if
> >> > you want. Since you are already using the slurmdbd what would be
> >> considered
> >> > the hard part is already over. Also, if you just want straight job
> >> records
> >> > no real setup is required.
> >> >
> >> > That would leave the jobcomp/script open for your other database.
> >> >
> >> > Danny
> >> >
> >> >> Yes, i guess in my expedience I wasn't clear. We're using the
> >> jobcomp
> >> >
> >> >> module with slurmdbd now, i want to continue using that, but also
> >> run
> >> >
> >> >> a script. effectively having two JobCompType parameters
> >> >
> >> >>
> >> >
> >> >> On Tue, Apr 19, 2011 at 11:37 AM, Jette, Moe <[email protected]>
> >> wrote:
> >> >
> >> >> > We do something like that today. Just configure
> >> >
> >> >> >
> >> >
> >> >> > JobCompType=jobcomp/script
> >> >
> >> >> > JobCompLoc=<pathname>
> >> >
> >> >> >
> >> >
> >> >> > Your script can easily write information into a database and do
> >> anything
> >> >> > else you want.
> >> >
> >> >> >
> >> >
> >> >> > ________________________________________
> >> >
> >> >> > From: [email protected] [owner-slurm-
> >> [email protected]] On
> >> >> > Behalf Of Michael Di Domenico [[email protected]]
> >> >
> >> >> > Sent: Tuesday, April 19, 2011 8:21 AM
> >> >
> >> >> > To: slurm-dev
> >> >
> >> >> > Subject: [slurm-dev] JobComp scripts
> >> >
> >> >> >
> >> >
> >> >> > Is it possible to have the JobComp logic insert it's records
> into
> >> a
> >> >
> >> >> > database, but also run a script for each job run?
> >> >
> >> >> >
> >> >
> >> >> >
> >> >
> >> >>
> >> >
> >> >
> >> >
> >
> >


Reply via email to