[Lustre-discuss] LustreError: 4267:0:(mds_reint.c:1596:mds_orphan_add_link()) LBUG

[Erik Gulliksson]

Hi list!

The following LBUG occured for us:

Mar 30 15:24:36 mdt1 kernel: LustreError:
4267:0:(mds_reint.c:1596:mds_orphan_add_link())
ASSERTION(inode->i_nlink == 1) failed:dir nlink == 0
Mar 30 15:24:36 mdt1 kernel: LustreError:
4267:0:(mds_reint.c:1596:mds_orphan_add_link()) LBUG
Mar 30 15:24:36 mdt1 kernel: Lustre:
4267:0:(linux-debug.c:185:libcfs_debug_dumpstack()) showing stack for
process 4267
Mar 30 15:24:36 mdt1 kernel: ll_mdt_12     R  running task       0
4267      1          4268  4266 (L-TLB)
Mar 30 15:24:36 mdt1 kernel:  0000003000000030 ffff81021d0036b0
ffff81021d0035d0 0000000000000001
Mar 30 15:24:36 mdt1 kernel:  0000000000000006 ffffffff803230cf
3830333233306366 3338333033333332
Mar 30 15:24:36 mdt1 kernel:  ffff81000107a000 0000000000000000
0000000000000000 ffffffff8035a14c
Mar 30 15:24:36 mdt1 kernel: Call Trace:
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8035a14c>] scrup+0x60/0xc5
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8035a1cd>] lf+0x1c/0x3a
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8035df08>] vt_console_print+0x213/0x228
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8035df08>] vt_console_print+0x213/0x228
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8028107a>] printk+0x4e/0x56
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8028107a>] printk+0x4e/0x56
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80297455>] kallsyms_lookup+0xe7/0x1af
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80297455>] kallsyms_lookup+0xe7/0x1af
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80261bbd>] printk_address+0x9f/0xac
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8035a14c>] scrup+0x60/0xc5
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80294719>] module_text_address+0x33/0x3b
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8028de3f>] kernel_text_address+0x1a/0x26
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8028de3f>] kernel_text_address+0x1a/0x26
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80261dc3>] show_trace+0x1f9/0x21f
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80261ee2>] _show_stack+0xe2/0xf1
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff88246b8a>]
:libcfs:lbug_with_loc+0x7a/0xc0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885d1cb1>]
:mds:mds_orphan_add_link+0x641/0x7e0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8854300d>]
:ldiskfs:__ldiskfs_journal_stop+0x2d/0x60
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8020c04f>] dnotify_parent+0x1c/0x6b
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8020c8c9>] dput+0x23/0x152
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885dce45>]
:mds:mds_reint_unlink+0x16c5/0x2410
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8020c8c9>] dput+0x23/0x152
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885d1599>]
:mds:mds_reint_rec+0x1d9/0x2b0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885f728c>]
:mds:mds_unlink_unpack+0x28c/0x3b0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8837c8bf>]
:ptlrpc:lustre_pack_reply_flags+0x7af/0x8c0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885c470a>] :mds:mds_reint+0x35a/0x420
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8837c9f9>]
:ptlrpc:lustre_pack_reply+0x29/0xb0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff88377d75>]
:ptlrpc:lustre_msg_get_opc+0x35/0xf0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff885cb345>] :mds:mds_handle+0x2475/0x4c20
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff802b5e3c>] free_block+0x53/0x131
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8027bcc2>]
find_busiest_group+0x20d/0x634
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff882dee68>]
:obdclass:class_handle2object+0xd8/0x160
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff883774e5>]
:ptlrpc:lustre_msg_get_conn_cnt+0x35/0xf0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff88380fda>]
:ptlrpc:ptlrpc_check_req+0x1a/0x110
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff883831a9>]
:ptlrpc:ptlrpc_server_handle_request+0x999/0x1040
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80263dde>] do_gettimeofday+0x50/0x94
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8824f276>]
:libcfs:lcw_update_time+0x16/0x100
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8027b0f4>] __wake_up_common+0x3e/0x68
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff883861ab>]
:ptlrpc:ptlrpc_main+0xe1b/0xfa0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff8027c74a>] default_wake_function+0x0/0xe
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80258820>] child_rip+0xa/0x12
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff88385390>] :ptlrpc:ptlrpc_main+0x0/0xfa0
Mar 30 15:24:36 mdt1 kernel:  [<ffffffff80258816>] child_rip+0x0/0x12

Have anyone seen this before? We are running 1.6.6 on kernel
2.6.18+2.6.18+lustre1.6.6+0.credativ.etch.1 (from Alioth .debs)


Best regards
Erik G

-- 
Erik Gulliksson, erik.gulliksson at diino.net
System Administrator, Diino AB
http://www.diino.com