Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-17704

network error in sanity-benchmark

    XMLWordPrintable

Details

    • Bug
    • Resolution: Unresolved
    • Minor
    • Lustre 2.16.0
    • None
    • None
    • 3
    • 9223372036854775807

    Description

      running sanity-benchmark I see this error in dbench subtest very often:

      [  137.575450] Lustre: 6926:0:(client.c:2340:ptlrpc_expire_one_request()) @@@ Request sent has failed due to network error: [sent 1712216405/real 1712216405]  req@ffff9d1ebb9d4980 x1795389298614272/t0(0) o104->lustre-MDT0001@0@lo:15/16 lens 328/224 e 0 to 1 dl 1712216421 ref 1 fl Rpc:ReXQU/0/ffffffff rc 0/-1 job:'' uid:4294967295 gid:4294967295
      

      bisection:

      COMMIT		TESTED	PASSED	FAILED		COMMIT DESCRIPTION
      3d635dd3f2	1	0	1	BAD	LU-17484 gss: reply error for SEC_CTX_INIT on wrong node
      02caf71707	1	0	1	BAD	LU-17258 socklnd: stop connecting on too many retries
      a6886dba0e	5	0	5	BAD	LU-17379 ptlrpc: fix check for callback discard
      adc60b8922	5	5	0	GOOD	LU-17000 utils: Use ssize_t to store return from sysconf()
      004af719b2	5	5	0	GOOD	LU-17507 build: Allow symlink in mofed default path
      099350d6e3	5	5	0	GOOD	LU-17505 socklnd: return NETWORK_TIMEOUT to LNet on ETIMEOUT
      dba4135556	5	5	0	GOOD	LU-17379 lnet: add LNetPeerDiscovered to LNet API
      

      Attachments

        Issue Links

          Activity

            People

              tappro Mikhail Pershin
              bzzz Alex Zhuravlev
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated: