Re: [HTCondor-users] Possible to have submit-implemented per-machine job limits?

[Todd Tannenbaum <tannenba@xxxxxxxxxxx>]

On 9/12/2016 12:31 PM, Michael V Pelletier wrote:
> Hi folks,
>
> We have a situation where a certain type of job has an adjunct service
> process which can only have on instance on a given machine, since it uses
> a static port number to provide its service to the job. It can't easily be
> reworked since it's designed to operate that way in a production
> environment. This means that one physical machine can only run one
> instance of that job.

Maybe you could run multiple instances of this adjunct job on one 
physical host by using a job universe that virtualizes the network 
environment (i.e. docker universe? vm universe?) ?

> I know I can set up a machine resource in the configuration for this
> purpose, assigning one "myservice" resource to each machine, and this
> would allow the job to specify "request_myservice = 1" and thus limit to
> one job per machine.
>
> What I'm wondering is if it's possible to use something in the job's
> requirements expression alone to accomplish this, rather than a
> server-side config customization. I'm using partitionable slots - I
> suspect that fact may make this a tricky problem to solve without startd
> configuration changes, because the partitionable slot would probably need
> information about what the dynamic slots are doing.

Doing what you want via setting up a custom machine resource (i.e. 
request_port777 = 1) is exactly what I'd suggest; scenarios like the 
above are why custom machine resources exist, since this really is a 
custom machine resource.  For instance, what if two different users both 
have an app that requires the same static slot?

But given that you cannot configure the execute nodes, perhaps your job 
requirements could look at the ChildRemoteUser attribute in the 
partitionable slot?  This attribute is a classad list of all the owners 
of dynamic slots on the machine.  You could probably leverage this so 
only one job submitted by you runs on each machine...

regards
Todd


> One similar thing I've done in the past was to steer jobs which could
> share a license checkout on the same machine by making a "condor_q" query
> from the script and turning it into a rank expression to favor machines
> already running that user's licensed jobs, but that requires, needless to
> say, a submit wrapper script which I'd like to avoid.
>
> I've also used SubmitterUserResourcesInUse, but that applies to the entire
> pool rather than to a single machine.
>
> Maybe there's some sort of trick in the new 8.4 submit syntax that could
> be applied here?
>
> Thanks for any suggestions you can offer!
>
>         -Michael Pelletier.
> _
>
>
>
> _______________________________________________
> HTCondor-users mailing list
> To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
> subject: Unsubscribe
> You can also unsubscribe by visiting
> https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users
>
> The archives can be found at:
> https://lists.cs.wisc.edu/archive/htcondor-users/


--
Todd Tannenbaum <tannenba@xxxxxxxxxxx> University of Wisconsin-Madison
Center for High Throughput Computing   Department of Computer Sciences
HTCondor Technical Lead                1210 W. Dayton St. Rm #4257
Phone: (608) 263-7132                  Madison, WI 53706-1685