Mailing List ArchivesPublic Access |
![[Computer Systems Lab]](https://www-auth.cs.wisc.edu/pics/csl_logo.gif){kind=logo}
|
Mailing List ArchivesPublic Access |
|
|
Hmmm… I should possibly add that I am running with partitionable slots. M From: HTCondor-users [mailto:htcondor-users-bounces@xxxxxxxxxxx]
On Behalf Of Marc Volovic Let's review this, again: marcv@htcondor[~]$
condor_config_val -dump | grep LIMIT CONCURRENCY_LIMIT_DEFAULT = 30 MARCV_LIMIT = 1 This should, to the best if my understanding, allow me to run no more than one job at any given one time. Note that the schedulers and the executors have
ConcurrencyLimits = Owner
STARTD_ATTRS = ConcurrencyLimits SUBMIT_ATTRS = ConcurrencyLimits Yet – when I submit 10 jobs, after a short while (a few minutes) – I have more than one running. Eventually, all 10 will run. What am I doing incorrectly? To wit: marcv@s1[~]$
condor_q -g | grep marcv marcv@s1[~]$
condor_q -r | grep marcv marcv@s1
[~]$
condor_userprio | grep marcv marcv@ezchip.com
0.59 1.00 0 15.51 0+00:01 marcv@s1[~]$
condor_q -- Submitter: s1.ezchip.com : <1.2.3.4:42660> : s1.ezchip.com ID OWNER SUBMITTED RUN_TIME ST PRI SIZE CMD 0 jobs; 0 completed, 0 removed, 0 idle, 0 running, 0 held, 0 suspended marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [1] 41818 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [2] 42013 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [3] 42180 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [4] 42463 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [5] 42779 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [6] 43033 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [7] 43390 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [8] 43590 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [9] 43736 marcv@s1[~]$
condor_run -a 'request_disk=10' -a 'request_memory=10' sleep 1d & [10] 43938 marcv@s1[~]$
condor_q -- Submitter: s1.ezchip.com : <1.2.3.4:42660> : s1.ezchip.com ID OWNER SUBMITTED RUN_TIME ST PRI SIZE CMD 110.0 marcv 8/11 10:44 0+00:01:09 R 0 0.0 .condor_run.41818 111.0 marcv 8/11 10:44 0+00:00:49 R 0 0.0 .condor_run.42013 112.0 marcv 8/11 10:44 0+00:00:29 R 0 0.0 .condor_run.42180 113.0 marcv 8/11 10:44 0+00:00:08 R 0 0.0 .condor_run.42463 114.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.42779 115.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.43033 116.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.43390 117.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.43590 118.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.43736 119.0 marcv 8/11 10:44 0+00:00:00 I 0 0.0 .condor_run.43938 10 jobs; 0 completed, 0 removed, 6 idle, 4 running, 0 held, 0 suspended From: HTCondor-users [mailto:htcondor-users-bounces@xxxxxxxxxxx]
On Behalf Of Marc Volovic J Thanks. I am acquainted with his page. That's the problem – I set up everything according to what I seem to read there, set my own personal concurrency limit
to 1 and – lo and behold – can run a gazillion jobs. M From: HTCondor-users [mailto:htcondor-users-bounces@xxxxxxxxxxx]
On Behalf Of Miha Ahronovitz
You are not the only one :-( Most of the people on this thread do not understand it, but they are afraid to admit it A blogger decided to publish his own discoveries Concurrency Limits: Protecting shared resources See if it inspires you. M --- --- --- --- --- --- --- --- --- --- --- --- --- Miha Ahronovitz Principal | Ahrono Associates Blog: http://my-inner-voice.blogspot.com/ c: 408
422 2757 tw: @myinnervoice --- --- --- --- --- --- --- --- --- --- --- --- --- On Sun, Aug 10, 2014 at 4:55 AM, Marc Volovic <marcv@xxxxxxxxxx> wrote: Hello, sorry for the latish response - we were on a weekend :-)...
|