[LU-7173] ldlm_lock_destroy_internal() LBUG encountered during 2.8 large scale testing

• Type: Bug
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.8.0
• Affects Version/s: Lustre 2.8.0
• Labels: patch
• Environment: Clients running Lustre-2.7.57 plus patches using an lustre 2.5.4 server back end.
• Severity: 3
• Assignee: Zhenyu Xu
• Reporter: James A Simmons

Description

While running a simulated user work load our MDS crashed due to the following:

<3>[ 5913.638287] LustreError: 15340:0:(ldlm_lock.c:371:ldlm_lock_destroy_internal()) ### lock still on resource ns: mdt-atlas1-MDT0000_UUID lock: ffff883fd37b3700/0xb9db2cef5d1e41
39 lrc: 3/0,0 mode: CR/CR res: [0x200252cc7:0x3:0x0].0 bits 0x8 rrc: 2 type: IBT flags: 0x50000000000000 nid: 9310@gni100 remote: 0xd3f59509a44985ef expref: 60 pid: 15340 timeout: 
0 lvb_type: 3
<0>[ 5913.674996] LustreError: 15340:0:(ldlm_lock.c:372:ldlm_lock_destroy_internal()) LBUG
<4>[ 5913.683915] Pid: 15340, comm: mdt03_052
<4>[ 5913.688327] 
<4>[ 5913.688327] Call Trace:
<4>[ 5913.692965]  [<ffffffffa0430895>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
<4>[ 5913.700886]  [<ffffffffa0430e97>] lbug_with_loc+0x47/0xb0 [libcfs]
<4>[ 5913.707968]  [<ffffffffa0707871>] ldlm_lock_destroy_internal+0x251/0x2c0 [ptlrpc]
<4>[ 5913.716573]  [<ffffffffa07092b5>] ldlm_lock_destroy+0x35/0x130 [ptlrpc]
<4>[ 5913.724118]  [<ffffffffa070a311>] ldlm_lock_enqueue+0x161/0x980 [ptlrpc]
<4>[ 5913.731760]  [<ffffffffa0733e9b>] ldlm_handle_enqueue0+0x51b/0x10c0 [ptlrpc]
<4>[ 5913.739797]  [<ffffffffa0d6f1a6>] mdt_enqueue+0x46/0xe0 [mdt]
<4>[ 5913.746362]  [<ffffffffa0d7401a>] mdt_handle_common+0x52a/0x1470 [mdt]
<4>[ 5913.753804]  [<ffffffffa0db0615>] mds_regular_handle+0x15/0x20 [mdt]
<4>[ 5913.761064]  [<ffffffffa0762f55>] ptlrpc_server_handle_request+0x385/0xc00 [ptlrpc]
<4>[ 5913.769866]  [<ffffffffa0442785>] ? lc_watchdog_touch+0x65/0x170 [libcfs]
<4>[ 5913.777632]  [<ffffffffa075b929>] ? ptlrpc_wait_event+0xa9/0x2d0 [ptlrpc]
<4>[ 5913.785377]  [<ffffffffa07656dd>] ptlrpc_main+0xaed/0x1930 [ptlrpc]
<4>[ 5913.792557]  [<ffffffffa0764bf0>] ? ptlrpc_main+0x0/0x1930 [ptlrpc]
<4>[ 5913.799689]  [<ffffffff8109e78e>] kthread+0x9e/0xc0
<4>[ 5913.805270]  [<ffffffff8100c28a>] child_rip+0xa/0x20
<4>[ 5913.810950]  [<ffffffff8109e6f0>] ? kthread+0x0/0xc0
<4>[ 5913.816622]  [<ffffffff8100c280>] ? child_rip+0x0/0x20

Issue Links

• is related to: LU-7535 mdt_intent_layout does not care about lock handles
• is related to: LU-6334 racer test_1: ldlm_lock_destroy_internal() LBUG