[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [HTCondor-users] Cannot sent jobs as Owner in WindowsOS



On 9/12/2018 5:02 AM, rb wrote:
> I would like to send and process the job as "owner".
> Not the default "condor-slot user" is procesing the job, but actually the person who is logged on the submitter and is sending the job.
> 
> For this we created a user "calibration*. This user is registered in our domain and has admin-permission on all machines (All win 10) connected to the pool.
> 
> For this I edited the config file on Submitter and Executing nodes:
> 
> [...]
> FILESYSTEM_DOMAIN = lgs-net.com
> UID_DOMAIN = lgs-net.com
> TRUST_UID_DOMAIN = true
> SOFT_UID_DOMAIN = true
> STARTER_ALLOW_RUNAS_OWNER = true
> [...]
> 
> 
> The submission files are having in addition following entry
> [...]
> Run_As_Owner = true
> [...]
> 
> 
> I also used "condor_store_cred add" on submitter and pool to store PW for user "calibration"
>  
> Still its not working!
> Jobs are created. Also .err and .out files. But they are not picked by Scheduler. Using "condor_q": No jobs in queue.
> 
> 
> Can someone give some hints?
> 

Did you do a condor_reconfig or restart HTCondor after changing the config settings on your execute and submit hosts?

Also I don't see anything in your config re your CREDD_HOST etc, as described in the Microsoft Windows chapter in the HTCondor Manual for executing jobs as the Submitting User... specifically I am looking at this section:
  http://htcondor.org/manual/v8.7/MicrosoftWindows.html#x75-5750008.2.4
Perhaps you want to re-read and follow the configuration examples in that part of the Manual.  

Some additional ideas / suggestions:

Are you running condor_submit as user "calibration" ?  What does "whoami" report before submitting the job?

Try submitting a very simple job and see if that runs as user "calibration".  I would suggest running "whoami.exe" with a job event log and see what happens.  For example -- 
   executable = whoami.exe
   output = test.out
   error = test.err
   log = test.log
   run_as_owner = true
   queue

and then take a look at test.out, test.err, test.log.

You say the job is successfully submitted but condor_q says no jobs in the queue... ??? what does "condor_q -allusers" say? Or is that because the job is quickly completing... what does condor_history say?

Re the below observations:  I am not the Windows expert, but I believe you should only need to run 'condor_store_cred add' on the submit node, which will then send the password (encrypted) and securely store it on the host running the condor_credd daemons.  The execute node will securely fetch the password as needed.

Hope the above helps,
Todd


> I made two observations:
> 1) I cannot use "condor_store_cred add" on executing machines. It returns an error "operation failed". Make sure you have WRITE permission onto this node. Although "WRITE = *" is set in all config files.
> 2) By default our Software adds "load_profile = true" in all submission files. Could this be a potential problem?
> 
> 
> 
> Best regards,
> Robert
> 
> 
> 
> 
> 
> 
> 
> -----------------------
> 
> -----------------------
> 
> _______________________________________________
> 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