Re: [HTCondor-users] GPUs_MONITOR resource usage

[Alexander Antoniades <sander@xxxxxxxxxxxx>]

I killed the process and cpu usage is the same.

USERÂ Â Â ÂPID %CPU %MEMÂ Â VSZÂ ÂRSS TTYÂ Â Â STAT STARTÂ ÂTIME COMMAND

condor Â26313 67.1 0.0 26460 6772 ?    S  15:39 14:44 GPUs_MONITOR

Here's a the summery of a quick strace if that helps.

root@gpu2:~# sort condor_gpu.txt |uniq -c

  793 getpid()                = 26313

 39637 ioctl(0, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x2a, 0x20), 0x7ffd1e446f30) = 0

  397 ioctl(0, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x2a, 0x20), 0x7ffd1e447110) = 0

   1 ioctl(0, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x2a, 0x20) <detached ...>

  396 ioctl(0, _IOC(_IOC_READ|_IOC_WRITE, 0x46, 0x4a, 0xb0), 0x7ffd1e4536a0) = 0

  Â49 nanosleep({tv_sec=0, tv_nsec=100000000}, NULL) = 0

   1 restart_syscall(<... resuming interrupted nanosleep ...>) = 0

  Â12 write(1, "SlotMergeConstraint = StringList"..., 135) = 135

   6 write(1, "SlotMergeConstraint = StringList"..., 136) = 136

   6 write(1, "SlotMergeConstraint = StringList"..., 139) = 139

Thanks,

Sander

On Wed, Jan 9, 2019 at 3:07 PM Todd L Miller <tlmiller@xxxxxxxxxxx> wrote:

> The cpu usage of GPUs_MONITOR seems a little high for sitting at idle.
> condor  2165 66.9 0.0 26460 6844 ?    R  Jan08 793:42
> GPUs_MONITOR

    I concur.

> Is the resource usage of the monitoring process normal?

    No. I think I only tested on machines with up to four GPUs, but
its CPU usage should always be trivial. I would just try killing it (this
will, of course, cause bogus GPU usage numbers to be reported) and seeing
it the problem re-occurs. If it does, I'll need to investigate.

- ToddM
_______________________________________________
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/