[LU-8743] client stuck in cl_inode_fini() Created: 20/Oct/16  Updated: 25/Oct/16  Resolved: 25/Oct/16

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.5.3
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Ned Bass Assignee: Zhenyu Xu
Resolution: Duplicate Votes: 0
Labels: llnl
Environment:

lustre-2.5.5-9chaos.2.ch6.x86_64
Linux quartz377 3.10.0-327.28.2.1chaos.ch6.x86_64 #1 SMP Wed Aug 3 15:09:48 PDT 2016 x86_64 x86_64 x86_64 GNU/Linux


Issue Links:
Duplicate
duplicates LU-8509 drop_caches hangs in cl_inode_fini() Resolved
Related
is related to LU-8509 drop_caches hangs in cl_inode_fini() Resolved
Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

Have a user process stuck with the following stacktrace.

[<ffffffffa11c2fdd>] cl_inode_fini+0x1ad/0x270 [lustre]
[<ffffffffa118511d>] ll_clear_inode+0x25d/0x9d0 [lustre]
[<ffffffffa11870d8>] ll_delete_inode+0x58/0x1d0 [lustre]
[<ffffffff812040b7>] evict+0xa7/0x170
[<ffffffff812049b5>] iput+0xf5/0x180
[<ffffffffa114de34>] ll_d_iput+0xf4/0x7f0 [lustre]
[<ffffffff811ffbe6>] dentry_kill+0x146/0x1b0
[<ffffffff811ffcac>] dput+0x5c/0xd0
[<ffffffff811e9aae>] __fput+0x18e/0x280
[<ffffffff811e9cee>] ____fput+0xe/0x10
[<ffffffff810a60b4>] task_work_run+0xb4/0xe0
[<ffffffff81016bb2>] do_notify_resume+0x92/0xb0
[<ffffffff8165dd3d>] int_signal+0x12/0x17
[<ffffffffffffffff>] 0xffffffffffffffff
(gdb) l *(cl_inode_fini+0x1ad)
0x75fdd is in cl_inode_fini (/usr/src/debug/lustre-2.5.5/lustre/lclient/lcommon_cl.c:1244).
1239
1240                    init_waitqueue_entry_current(&waiter);
1241                    add_wait_queue(&bkt->lsb_marche_funebre, &waiter);
1242
1243                    while (1) {
1244                            set_current_state(TASK_UNINTERRUPTIBLE);
1245                            if (cfs_atomic_read(&header->loh_ref) == 1)
1246                                    break;
1247                            waitq_wait(&waiter, TASK_UNINTERRUPTIBLE);
1248                    }
(gdb)


 Comments   
Comment by Peter Jones [ 20/Oct/16 ]

Bobijam

Could you please advise on this one?

Thanks

Peter

Comment by Zhenyu Xu [ 25/Oct/16 ]

Does this chaos branch contains #7893 and #10503 patches?

Comment by Peter Jones [ 25/Oct/16 ]

Yes both of those are in 2.8.0 itself.

Comment by Zhenyu Xu [ 25/Oct/16 ]

http://review.whamcloud.com/#/c/23316/ then contains the fix.

Comment by Peter Jones [ 25/Oct/16 ]

ok so this is believed to be a duplicate of LU-8509, which fix has not been picked up yet but is in progress

Generated at Sat Feb 10 02:20:10 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.