Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-9521

obdfilter-survey exits with error messages

    Details

    • Type: Bug
    • Status: Resolved
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: Lustre 2.10.0
    • Labels:
    • Environment:
      Lustre - 2.9.57_66_gb6d5b5b
    • Severity:
      3
    • Rank (Obsolete):
      9223372036854775807

      Description

      obdfilter-survey seems to run to completion but prints some error messages and exits with a non-zero error code as shown below.

      [root@hsmdev tests]# TERM=linux thrlo=1 thrhi=32 nobjlo=1 nobjhi=1 size=1310 rszlo=1024 rszhi=1024 tests_str="write read" targets="lustre-OST0000" /usr/bin/obdfilter-survey
      
      Wed May 17 17:55:58 IST 2017 Obdfilter-survey for case=disk from hsmdev
      ost  1 sz  1341440K rsz 1024K obj    1 thr    1 write  126.93 [  11.00,  210.94] read  488.46 [ 220.75,  220.75] 
      ost  1 sz  1341440K rsz 1024K obj    1 thr    2 write  137.33 [  14.99,  241.94] read  664.90             SHORT 
      ost  1 sz  1339392K rsz 1024K obj    1 thr    4 write  175.87 [  16.99,  456.63] read  729.53             SHORT 
      ost  1 sz  1335296K rsz 1024K obj    1 thr    8 write  192.75 [  12.98,  493.87] read  706.75             SHORT 
      ost  1 sz  1327104K rsz 1024K obj    1 thr   16 write  213.86 [  18.95,   73.88] read  616.42             SHORT 
      ost  1 sz  1310720K rsz 1024K obj    1 thr   32 write  221.85 [  23.97,  339.91] read  607.29             SHORT 
      /usr/bin/iokit-libecho: line 236: 35781 Killed                  remote_shell $host "vmstat 5 >> $host_vmstatf" &>/dev/null
      done!
      No device found for name lustre-OST0000_ecc: Invalid argument
      [root@hsmdev tests]# echo $?
      1
      [root@hsmdev tests]#
      

      Following change helps fix it.

      diff --git a/lustre-iokit/obdfilter-survey/iokit-libecho b/lustre-iokit/obdfilter-survey/iokit-libecho
      index 58e6b5a..81170a3 100644
      --- a/lustre-iokit/obdfilter-survey/iokit-libecho
      +++ b/lustre-iokit/obdfilter-survey/iokit-libecho
      @@ -217,6 +217,7 @@ cleanup_network () {
       
       # do cleanup and exit.
       cleanup () {
      +	trap 0
       	local exit_status=$1
       	local host
      
      [root@hsmdev tests]# TERM=linux thrlo=1 thrhi=32 nobjlo=1 nobjhi=1 size=1310 rszlo=1024 rszhi=1024 tests_str="write read" targets="lustre-OST0000" /usr/bin/obdfilter-survey
      
      Wed May 17 18:06:10 IST 2017 Obdfilter-survey for case=disk from hsmdev
      ost  1 sz  1341440K rsz 1024K obj    1 thr    1 write  127.07 [  23.99,  218.95] read  562.72 [ 654.52,  654.52] 
      ost  1 sz  1341440K rsz 1024K obj    1 thr    2 write  134.74 [  11.98,  222.94] read  732.33             SHORT 
      ost  1 sz  1339392K rsz 1024K obj    1 thr    4 write  177.41 [  18.00,  434.88] read  659.88             SHORT 
      ost  1 sz  1335296K rsz 1024K obj    1 thr    8 write  209.18 [  14.99,  729.80] read  691.23             SHORT 
      ost  1 sz  1327104K rsz 1024K obj    1 thr   16 write  220.12 [  14.99,  157.87] read  610.35             SHORT 
      ost  1 sz  1310720K rsz 1024K obj    1 thr   32 write  227.72 [  27.93,  362.86] read  608.37             SHORT 
      /usr/bin/iokit-libecho: line 236: 47023 Killed                  remote_shell $host "vmstat 5 >> $host_vmstatf" &>/dev/null
      done!
      [root@hsmdev tests]# echo $?
      0
      [root@hsmdev tests]#
      

        Attachments

          Activity

            People

            • Assignee:
              wc-triage WC Triage
              Reporter:
              parinay parinay v kondekar (Inactive)
            • Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: