[LU-10662] LBUG: (statahead.c:1262:ll_deauthorize_statahead()) ASSERTION( lli->lli_opendir_pid != 0 )

• Type: Improvement
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.11.0
• Assignee: Ben Evans (Inactive)
• Reporter: Ben Evans (Inactive)

Description

=== NODE FAILURES ===

LUS: LBUG-ASSERT(lli->lli_opendir_pid != 0) in ll_deauthorize_statahead (statahead.c:1262)
First hit Cname # hits Apids/Roles
----------------- ------------- ------ ---------------
$"-16/04/03 00:35:36 c0-0c2s7n0 1

$ cdumps/160403064850/c0-0c2s7n0-1604030648.cdump
" xt/console-buffers/c0-0c2s7n0.__log_buf

$ = node memory dump available

Node is run out memory:

> 2016-04-03T00:35:36.841762-05:00 c0-0c2s7n0 Killed process 12138 (stressapptest) apid 8953 total-vm:32659880kB, anon-rss:11876800kB, file-rss:0kB
> 2016-04-03T00:35:36.841775-05:00 c0-0c2s7n0 Memory cgroup out of memory: Killed 20 processes sharing cpu group with pid 12138.

ll_file_data slab can't be allocated. Lustre panics trying to handle the ENOMEM.

> 2016-04-03T00:35:36.841499-05:00 c0-0c2s7n0 cache: ll_file_data(29:step_0), object size: 256, order: 0
> 2016-04-03T00:35:36.841507-05:00 c0-0c2s7n0 node 0: slabs: 14/14, objs: 210/210, free: 0
> 2016-04-03T00:35:36.841515-05:00 c0-0c2s7n0 node 1: slabs: 16/16, objs: 240/240, free: 0
> 2016-04-03T00:35:36.841555-05:00 c0-0c2s7n0 LustreError: 12127:0:(statahead.c:1262:ll_deauthorize_statahead()) ASSERTION( lli->lli_opendir_pid != 0 ) failed:
> 2016-04-03T00:35:36.841569-05:00 c0-0c2s7n0 LustreError: 12127:0:(statahead.c:1262:ll_deauthorize_statahead()) LBUG
> 2016-04-03T00:35:36.841583-05:00 c0-0c2s7n0 Pid: 12127, comm: growfiles
> 2016-04-03T00:35:36.841597-05:00 c0-0c2s7n0 Call Trace:
> 2016-04-03T00:35:36.841610-05:00 c0-0c2s7n0 [<ffffffff81005d71>] try_stack_unwind+0x191/0x1a0
> 2016-04-03T00:35:36.841621-05:00 c0-0c2s7n0 [<ffffffff810047eb>] dump_trace+0x8b/0x350
> 2016-04-03T00:35:36.841635-05:00 c0-0c2s7n0 [<ffffffffa0267813>] libcfs_debug_dumpstack+0x53/0x80 [libcfs]
> 2016-04-03T00:35:36.841647-05:00 c0-0c2s7n0 [<ffffffffa0267da5>] lbug_with_loc+0x45/0xc0 [libcfs]
> 2016-04-03T00:35:36.841661-05:00 c0-0c2s7n0 [<ffffffffa08e834a>] ll_deauthorize_statahead+0x17a/0x180 [lustre]
> 2016-04-03T00:35:36.841671-05:00 c0-0c2s7n0 [<ffffffffa0897307>] ll_file_open+0x977/0xdd0 [lustre]
> 2016-04-03T00:35:36.841683-05:00 c0-0c2s7n0 [<ffffffff8118005e>] do_dentry_open.isra.17+0x1de/0x280
> 2016-04-03T00:35:36.841697-05:00 c0-0c2s7n0 [<ffffffff8118011e>] finish_open+0x1e/0x30
> 2016-04-03T00:35:36.841710-05:00 c0-0c2s7n0 [<ffffffffa08cad34>] ll_atomic_open+0x594/0x11b0 [lustre]
> 2016-04-03T00:35:36.841722-05:00 c0-0c2s7n0 [<ffffffff81190c29>] do_last+0x899/0x11e0
> 2016-04-03T00:35:36.841735-05:00 c0-0c2s7n0 [<ffffffff8119162b>] path_openat+0xbb/0x640
> 2016-04-03T00:35:36.841748-05:00 c0-0c2s7n0 [<ffffffff811929aa>] do_filp_open+0x3a/0x90
> 2016-04-03T00:35:36.841789-05:00 c0-0c2s7n0 [<ffffffff811816d8>] do_sys_open+0x128/0x220
> 2016-04-03T00:35:36.841802-05:00 c0-0c2s7n0 [<ffffffff811817ee>] SyS_open+0x1e/0x20
> 2016-04-03T00:35:36.841814-05:00 c0-0c2s7n0 [<ffffffff8149fab2>] system_call_fastpath+0x16/0x1b
> 2016-04-03T00:35:36.841827-05:00 c0-0c2s7n0 [<0000000020022f50>] 0x20022f50
> 2016-04-03T00:35:36.841835-05:00 c0-0c2s7n0 Kernel panic - not syncing: LBUG