[HTCondor-users] Weirdness with cpususage

[Jeff Templon <templon@xxxxxxxxx>]

Hi,

I was trying to use Cpususage in a meaningful way, but itâs not working.  See below (done-cpususage.cpf included at end of mail):

condor_history  -completedsince $(date -d "7 days ago" +"%s") -constraint 'RemoteUserCpu > 10*RemoteWallClockTime' -print-format ~templon/done-cpususage.cpf -wide:132 | head -20

JOB_ID    Username    CMD              Finished   CPUS   CPUS used     RAM      MEM      ST   CPU_TIME     WALL_TIME  WorkerNode

703590.0 admayol     run_63.sh          10/25 09:35  32 0.435   1.9 GB     2.9 GB C     2+02:02:22      3:47:09 wn-lot-054

703595.0 admayol     run_68.sh          10/25 09:16  32 0.594   2.0 GB     3.1 GB C     1+01:30:54      1:57:07 wn-lot-044

703593.0 admayol     run_66.sh          10/25 08:42  32 14.76   2.0 GB     2.9 GB C     1+00:22:37      1:54:23 wn-lot-048

703592.0 admayol     run_65.sh          10/25 08:18  32 12.52   1.9 GB     2.9 GB C     1+00:09:40      1:51:30 wn-lot-052

696992.0 roystege    pineko theory -c / 10/25 07:18  32 0.997  11.5 GB    11.9 GB C    22+11:17:47     23:34:58 wn-lot-044

696632.0 roystege    pineko theory -c / 10/25 06:51  32 0.997  11.8 GB    14.3 GB C    31+12:35:56   1+09:11:12 wn-lot-056

703589.0 admayol     run_62.sh          10/25 06:26  32 13.76   1.9 GB     2.6 GB C       20:39:48      1:37:30 wn-lot-052

696917.0 roystege    pineko theory -c / 10/25 06:00  32 0.999  11.7 GB    11.9 GB C    22+21:33:26     23:55:36 wn-lot-047

703586.0 admayol     run_59.sh          10/25 04:48  32 13.22   1.8 GB     2.6 GB C       19:34:31      1:31:57 wn-lot-052

696842.0 roystege    pineko theory -c / 10/25 04:33  32 0.997  11.6 GB    11.9 GB C    23+01:42:46   1+00:12:05 wn-lot-048

703585.0 admayol     run_58.sh          10/25 04:01  32 14.85   2.2 GB     2.9 GB C     1+01:50:47      1:55:44 wn-lot-054

696767.0 roystege    pineko theory -c / 10/25 03:41  32 1.000  12.5 GB    14.3 GB C    24+22:23:19   1+01:27:43 wn-knek-002

703584.0 admayol     run_57.sh          10/25 03:16  32 0.610   2.1 GB     3.1 GB C     1+05:00:05      2:14:22 wn-lot-052

The CPUS used column is Cpususage, and should be according to the docs, CPU_TIME divided by WALL_TIME. In the output, sometimes it is, and sometimes it isnât. 

How is this possible?

JT

done-cpususage.cpf:

SELECT NOSUMMARY

   ClusterId                      AS  JOB_ID      PRINTAS JOB_ID

   Owner                          AS '   Username'

   join(" ",split(Cmd,"/")[size(split(Cmd,"/"))-1], Args)  AS '   CMD' WIDTH -18

   CompletionDate                 AS ' Finished ' PRINTAS DATE

   CpusProvisioned                AS ' CPUS '  WIDTH 3

   CpusUsage                      AS ' CPUS used ' WIDTH 5

#   MemoryProvisioned              AS ' MEMREQ' PRINTAS READABLE_MB

   ResidentSetSize_RAW            AS '   RAM' PRINTAS READABLE_KB  WIDTH 8

   ImageSize                      AS '   MEM' PRINTAS READABLE_KB  WIDTH 10

   JobStatus                      AS " ST"        PRINTAS JOB_STATUS

   interval(RemoteUserCpu)        AS  "  CPU_TIME"    WIDTH 12

   interval(RemoteWallClockTime)  AS  "  WALL_TIME"   WIDTH 12

   split(splitSlotName(LastRemoteHost)[1], ".")[0]  AS  "WorkerNode" WIDTH -12