Uploaded image for project: 'Lustre'
  1. Lustre
  2. LU-16907

sanity test_123f: crashed MDS with Max IOV exceeded: 257 should be < 256

    XMLWordPrintable

Details

    • Bug
    • Resolution: Fixed
    • Minor
    • Lustre 2.17.0
    • None
    • None
    • 3
    • 9223372036854775807

    Description

      This issue was created by maloo for Andreas Dilger <adilger@whamcloud.com>

      This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/85009820-d429-41a1-8948-90b2f66d7f02

      test_123f failed with the following error:

      onyx-37vm9 crashed during sanity test_123f
      
      LNetError: 191018:0:(socklnd_cb.c:1036:ksocknal_send()) ASSERTION( tx->tx_nkiov <= 256 ) failed: 
      LNetError: 191018:0:(socklnd_cb.c:1036:ksocknal_send()) LBUG
      Pid: 191018, comm: mdt_out00_001 4.18.0-425.10.1.el8_lustre.x86_64 #1 SMP Wed May 3 16:22:26 UTC 2023
      Call Trace TBD:
       libcfs_call_trace+0x6f/0xa0 [libcfs]
       lbug_with_loc+0x3f/0x70 [libcfs]
       ksocknal_send+0x27a/0x320 [ksocklnd]
       lnet_ni_send+0x4c/0xe0 [lnet]
       lnet_send+0xae/0x1e0 [lnet]
       LNetPut+0x318/0x940 [lnet]
       ptl_send_buf+0x208/0x5a0 [ptlrpc]
       ptlrpc_send_reply+0x2ad/0x8d0 [ptlrpc]
       target_send_reply+0x328/0x7d0 [ptlrpc]
       tgt_request_handle+0xe85/0x1920 [ptlrpc]
       ptlrpc_server_handle_request+0x31d/0xbc0 [ptlrpc]
       ptlrpc_main+0xc52/0x1510
      

      Test session details:
      clients: https://build.whamcloud.com/job/lustre-reviews/94706 - 5.4.0-131-generic
      servers: https://build.whamcloud.com/job/lustre-reviews/94706 - 4.18.0-425.10.1.el8_lustre.x86_64

      Have seen this about 10 times since 2023-05-09, after patch https://review.whamcloud.com/46540 "LU-15550 ptlrpc: retry mechanism for overflowed batched RPCs" landed, but I'm not sure if it is directly related.

      VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
      sanity test_123f - onyx-37vm9 crashed during sanity test_123f

      Attachments

        Issue Links

          Activity

            People

              qian_wc Qian Yingjin
              maloo Maloo
              Votes:
              0 Vote for this issue
              Watchers:
              9 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: