[LU-1429] LBUG while unmounting client

• Type: Bug
• Resolution: Duplicate
• Priority: Minor

Environment

Lustre servers are running 2.6.32-220.el6, with Lustre 2.1.1.rc4.
Lustre clients are running 2.6.38.2, with special code created for this release, with http://review.whamcloud.com/#change,2170. (patch 8)

• Severity: 3
• Assignee: Lai Siyao
• Reporter: Roger Spellman (Inactive)

Description

The customer reports:

I was unmounting all of the lustre clients this morning in preparation of upgrading the system to test a bug fix when one of the clients panic'd:

2012-05-21 14:19:56 +0000 [3286924.778603] LustreError: 21993:0:(ldlm_request.c:1170:ldlm_cli_cancel_req()) Got rc -108 from cancel RPC: canceling anyway
2012-05-21 14:19:56 +0000 [3286924.836572] LustreError: 21993:0:(ldlm_request.c:1796:ldlm_cli_cancel_list()) ldlm_cli_cancel_list: -108
2012-05-21 14:19:56 +0000 [3286924.885902] LustreError: 21993:0:(ldlm_request.c:1170:ldlm_cli_cancel_req()) Got rc -108 from cancel RPC: canceling anyway
2012-05-21 14:19:56 +0000 [3286924.942180] LustreError: 21993:0:(ldlm_request.c:1796:ldlm_cli_cancel_list()) ldlm_cli_cancel_list: -108
2012-05-21 14:19:56 +0000 [3286925.029419] LustreError: 21993:0:(ldlm_lock.c:1749:ldlm_lock_cancel()) ### lock still has references ns: xxxxxxx-MDT0000-mdc-ffff880baa687800 lock: ffff88179ce22ac0/0xf44cc339ed65dcfe lrc: 4/1,0 mode: PR/PR res: 8593723616/109284 bits 0x3 rrc: 2 type: IBT flags: 0x22002890 remote: 0x45adc59ebc8ce531 expref: -99 pid: 22254 timeout: 0
2012-05-21 14:19:56 +0000 [3286925.178583] LustreError: 21993:0:(ldlm_lock.c:1750:ldlm_lock_cancel()) LBUG
2012-05-21 14:19:56 +0000 [3286925.214528] Pid: 21993, comm: umount
2012-05-21 14:19:56 +0000 [3286925.233175]
2012-05-21 14:19:56 +0000 [3286925.233176] Call Trace:
2012-05-21 14:19:56 +0000 [3286925.254280] [<ffffffffa04da875>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
2012-05-21 14:19:56 +0000 [3286925.290628] [<ffffffffa04dada7>] lbug_with_loc+0x47/0xc0 [libcfs]
2012-05-21 14:19:56 +0000 [3286925.322003] [<ffffffffa06ff50a>] ldlm_lock_cancel+0x13a/0x140 [ptlrpc]
2012-05-21 14:19:56 +0000 [3286925.356112] [<ffffffffa07158ff>] ldlm_cli_cancel_local+0xbf/0x310 [ptlrpc]
2012-05-21 14:19:56 +0000 [3286925.392166] [<ffffffffa0719542>] ldlm_cli_cancel+0x62/0x340 [ptlrpc]
2012-05-21 14:19:56 +0000 [3286925.424934] [<ffffffffa0706455>] cleanup_resource+0x175/0x2d0 [ptlrpc]
2012-05-21 14:19:56 +0000 [3286925.460116] [<ffffffff8112b8de>] ? __free_slab+0xce/0x180
2012-05-21 14:19:57 +0000 [3286925.488137] [<ffffffffa07065da>] ldlm_resource_clean+0x2a/0x50 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.524533] [<ffffffffa04e8cce>] cfs_hash_for_each_relax+0x18e/0x350 [libcfs]
2012-05-21 14:19:57 +0000 [3286925.562352] [<ffffffffa07065b0>] ? ldlm_resource_clean+0x0/0x50 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.597063] [<ffffffffa07065b0>] ? ldlm_resource_clean+0x0/0x50 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.632590] [<ffffffffa04eb6a8>] cfs_hash_for_each_nolock+0x88/0x180 [libcfs]
2012-05-21 14:19:57 +0000 [3286925.671295] [<ffffffffa0703a39>] ldlm_namespace_cleanup+0x29/0x80 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.706770] [<ffffffffa07048c7>] __ldlm_namespace_free+0x57/0x470 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.743179] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.782004] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.819882] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.858343] [<ffffffffa04eb6b0>] ? cfs_hash_for_each_nolock+0x90/0x180 [libcfs]
2012-05-21 14:19:57 +0000 [3286925.896777] [<ffffffffa070542a>] ldlm_namespace_free_prior+0x7a/0x190 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.933682] [<ffffffffa070d8b5>] client_disconnect_export+0x1d5/0x350 [ptlrpc]
2012-05-21 14:19:57 +0000 [3286925.972642] [<ffffffffa0c4e615>] lmv_disconnect+0x505/0x9a0 [lmv]
2012-05-21 14:19:57 +0000 [3286926.004722] [<ffffffffa0b8651e>] client_common_put_super+0x4fe/0xb30 [lustre]
2012-05-21 14:19:57 +0000 [3286926.041470] [<ffffffffa0b86c33>] ll_put_super+0xe3/0x280 [lustre]
2012-05-21 14:19:57 +0000 [3286926.073663] [<ffffffff81151ab6>] ? destroy_inode+0x36/0x60
2012-05-21 14:19:57 +0000 [3286926.102214] [<ffffffff811524fb>] ? dispose_list+0xdb/0x100
2012-05-21 14:19:57 +0000 [3286926.131705] [<ffffffff8113b1c2>] generic_shutdown_super+0x72/0x100
2012-05-21 14:19:57 +0000 [3286926.163750] [<ffffffff8113b2e6>] kill_anon_super+0x16/0x60
2012-05-21 14:19:57 +0000 [3286926.193020] [<ffffffffa05d8ef6>] lustre_kill_super+0x36/0x50 [obdclass]
2012-05-21 14:19:57 +0000 [3286926.227717] [<ffffffff8113ba45>] deactivate_locked_super+0x45/0x60
2012-05-21 14:19:57 +0000 [3286926.259521] [<ffffffff8113c9aa>] deactivate_super+0x4a/0x70
2012-05-21 14:19:57 +0000 [3286926.289284] [<ffffffff811567af>] mntput_no_expire+0x11f/0x190
2012-05-21 14:19:57 +0000 [3286926.318995] [<ffffffff81157cc8>] sys_umount+0x78/0x3c0
2012-05-21 14:19:57 +0000 [3286926.346323] [<ffffffff8100bfc2>] system_call_fastpath+0x16/0x1b
2012-05-21 14:19:57 +0000 [3286926.376895]
2012-05-21 14:19:57 +0000 [3286926.385435] Kernel panic - not syncing: LBUG
2012-05-21 14:19:57 +0000 [3286926.408223] Pid: 21993, comm: umount Tainted: G W 2.6.38.2-ts4 #11
2012-05-21 14:19:57 +0000 [3286926.443832] Call Trace:
2012-05-21 14:19:57 +0000 [3286926.458387] [<ffffffff8145a373>] ? panic+0x91/0x19c
2012-05-21 14:19:58 +0000 [3286926.483917] [<ffffffffa04dae0b>] ? lbug_with_loc+0xab/0xc0 [libcfs]
2012-05-21 14:19:58 +0000 [3286926.517230] [<ffffffffa06ff50a>] ? ldlm_lock_cancel+0x13a/0x140 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.551815] [<ffffffffa07158ff>] ? ldlm_cli_cancel_local+0xbf/0x310 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.589084] [<ffffffffa0719542>] ? ldlm_cli_cancel+0x62/0x340 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.623771] [<ffffffffa0706455>] ? cleanup_resource+0x175/0x2d0 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.658412] [<ffffffff8112b8de>] ? __free_slab+0xce/0x180
2012-05-21 14:19:58 +0000 [3286926.687756] [<ffffffffa07065da>] ? ldlm_resource_clean+0x2a/0x50 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.723131] [<ffffffffa04e8cce>] ? cfs_hash_for_each_relax+0x18e/0x350 [libcfs]
2012-05-21 14:19:58 +0000 [3286926.762757] [<ffffffffa07065b0>] ? ldlm_resource_clean+0x0/0x50 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.798172] [<ffffffffa07065b0>] ? ldlm_resource_clean+0x0/0x50 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.832517] [<ffffffffa04eb6a8>] ? cfs_hash_for_each_nolock+0x88/0x180 [libcfs]
2012-05-21 14:19:58 +0000 [3286926.870484] [<ffffffffa0703a39>] ? ldlm_namespace_cleanup+0x29/0x80 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.907478] [<ffffffffa07048c7>] ? __ldlm_namespace_free+0x57/0x470 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.943632] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286926.982120] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286927.020929] [<ffffffffa07188d0>] ? ldlm_cli_hash_cancel_unused+0x0/0x70 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286927.058688] [<ffffffffa04eb6b0>] ? cfs_hash_for_each_nolock+0x90/0x180 [libcfs]
2012-05-21 14:19:58 +0000 [3286927.097335] [<ffffffffa070542a>] ? ldlm_namespace_free_prior+0x7a/0x190 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286927.136255] [<ffffffffa070d8b5>] ? client_disconnect_export+0x1d5/0x350 [ptlrpc]
2012-05-21 14:19:58 +0000 [3286927.174598] [<ffffffffa0c4e615>] ? lmv_disconnect+0x505/0x9a0 [lmv]
2012-05-21 14:19:58 +0000 [3286927.207396] [<ffffffffa0b8651e>] ? client_common_put_super+0x4fe/0xb30 [lustre]
2012-05-21 14:19:58 +0000 [3286927.245652] [<ffffffffa0b86c33>] ? ll_put_super+0xe3/0x280 [lustre]
2012-05-21 14:19:58 +0000 [3286927.278089] [<ffffffff81151ab6>] ? destroy_inode+0x36/0x60
2012-05-21 14:19:58 +0000 [3286927.306992] [<ffffffff811524fb>] ? dispose_list+0xdb/0x100
2012-05-21 14:19:58 +0000 [3286927.335308] [<ffffffff8113b1c2>] ? generic_shutdown_super+0x72/0x100
2012-05-21 14:19:58 +0000 [3286927.368598] [<ffffffff8113b2e6>] ? kill_anon_super+0x16/0x60
2012-05-21 14:19:58 +0000 [3286927.398226] [<ffffffffa05d8ef6>] ? lustre_kill_super+0x36/0x50 [obdclass]
2012-05-21 14:19:58 +0000 [3286927.433461] [<ffffffff8113ba45>] ? deactivate_locked_super+0x45/0x60
2012-05-21 14:19:59 +0000 [3286927.467365] [<ffffffff8113c9aa>] ? deactivate_super+0x4a/0x70
2012-05-21 14:19:59 +0000 [3286927.497891] [<ffffffff811567af>] ? mntput_no_expire+0x11f/0x190
2012-05-21 14:19:59 +0000 [3286927.529535] [<ffffffff81157cc8>] ? sys_umount+0x78/0x3c0
2012-05-21 14:19:59 +0000 [3286927.557320] [<ffffffff8100bfc2>] ? system_call_fastpath+0x16/0x1b