* Re: PROBLEM: 3.6.0 kernel BUG at fs/dcache.c:967 during shutdown / restart
2012-10-09 7:54 ` PROBLEM: 3.6.0 kernel BUG at fs/dcache.c:967 during shutdown / restart Jan Kara
@ 2012-10-09 8:05 ` Stanislav Kinsbursky
0 siblings, 0 replies; 2+ messages in thread
From: Stanislav Kinsbursky @ 2012-10-09 8:05 UTC (permalink / raw)
To: Jan Kara
Cc: Neil Salstrom, linux-kernel@vger.kernel.org,
linux-nfs@vger.kernel.org, J. Bruce Fields
09.10.2012 11:54, Jan Kara пишет:
> On Mon 08-10-12 17:10:52, Neil Salstrom wrote:
>> On 10/08/2012 10:14 AM, Jan Kara wrote:
>>> On Sat 06-10-12 10:20:26, Neil Salstrom wrote:
>>>> I've not submitted a kernel bug before but I've read the bug
>>>> reporting pages. I'll try to do my best and if you need more
>>>> information please let me know. Please feel free to cc me on the
>>>> answer or if you have questions.
>>>>
>>>> I'm using Linux Mint 13 (64 bit) on AMD hardware with a self
>>>> compiled 3.6.0 kernel for my MythTV HTPC.
>>>>
>>>> Kernel version: Linux version 3.6.0 (gcc version 4.6.3
>>>> (Ubuntu/Linaro 4.6.3-1ubuntu5) ) #1 SMP Tue Oct 2 15:34:46 PDT 2012
>>>>
>>>> CPU: model name : AMD A6-3500 APU with Radeon(tm) HD Graphics
>>>>
>>>> If you need any other system variables please let me know and I can
>>>> get those as well.
>>>>
>>>> Upon shutdown or restart (from either the hard button on the case or
>>>> a "sudo shutdown -r now" the system crashes and does not actually
>>>> shut down. I can ssh in to the system and a dmesg shows the
>>>> following:
>>> There should be a line like:
>>> BUG: Dentry ... still in use (%d) [unmount of %s %s]
>>>
>>> just before the oops. Can you please post that one as well? Also I see that
>>> you are using nvidia module which taints the kernel. Can you try
>>> reproducing the issue without the module loaded?
>>>
>>> Honza
>>
>> Here is the full output with no nvidia module. I also have the full
>> dmeg from start to finish if you would like it.
> Thanks for the new report. Since the use count of the problematic dentry
> is 1, I guess this is really some bug in rpc_pipefs. Bruce, any idea?
>
Hi.
Could you, please, give more details about your configuration.
Currently, PipeFS is mounted on SUNRPC module install.
Would be great to understand, what process tries to umount it and when.
Also, do you have NFS mounts on the node, then Pipefs is umounting?
Is systemd is active on your node?
Could you trigger this panic manually?
> Honza
>
>> [ 237.186167] BUG: Dentry ffff880118772540{i=6e,n=info} still in
>> use (1) [unmount of rpc_pipefs rpc_pipefs]
>> [ 237.186179] ------------[ cut here ]------------
>> [ 237.186181] kernel BUG at fs/dcache.c:967!
>> [ 237.186182] invalid opcode: 0000 [#1] SMP
>> [ 237.186185] Modules linked in: nls_utf8 udf crc_itu_t nfsv4
>> binfmt_misc nfsd nfs_acl auth_rpcgss nfs fscache lockd sunrpc
>> snd_hda_codec_hdmi snd_hda_codec_realtek xfs snd_hda_intel
>> snd_hda_codec snd_seq_midi snd_hwdep snd_rawmidi snd_pcm psmouse
>> snd_seq_midi_event rc_imon_mce serio_raw snd_seq imon rc_core
>> snd_timer snd_seq_device snd soundcore snd_page_alloc f71882fg r8169
>> ahci libahci
>> [ 237.186203] CPU 2
>> [ 237.186206] Pid: 3063, comm: umount Not tainted 3.6.0 #1 System
>> manufacturer System Product Name/F1A55-M LX PLUS
>> [ 237.186208] RIP: 0010:[<ffffffff8116c606>] [<ffffffff8116c606>]
>> shrink_dcache_for_umount_subtree+0x1e6/0x1f0
>> [ 237.186214] RSP: 0018:ffff8801181afd98 EFLAGS: 00010296
>> [ 237.186215] RAX: 000000000000005d RBX: ffff880118772540 RCX:
>> 00000000000014bd
>> [ 237.186217] RDX: 0000000000003256 RSI: 0000000000000046 RDI:
>> ffffffff81bbf9bc
>> [ 237.186218] RBP: ffff8801181afdb8 R08: 000000000000000a R09:
>> 0000000000000000
>> [ 237.186219] R10: 0000000000000383 R11: 0000000000000382 R12:
>> ffff88011873b9c0
>> [ 237.186220] R13: ffff8801177a2b80 R14: ffff8801177a2bb0 R15:
>> ffff88011287d800
>> [ 237.186222] FS: 00007f67049b1800(0000) GS:ffff88011ed00000(0000)
>> knlGS:0000000000000000
>> [ 237.186223] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
>> [ 237.186225] CR2: 00007f3e1de80000 CR3: 0000000118a18000 CR4:
>> 00000000000007e0
>> [ 237.186226] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
>> 0000000000000000
>> [ 237.186227] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
>> 0000000000000400
>> [ 237.186229] Process umount (pid: 3063, threadinfo
>> ffff8801181ae000, task ffff880110fbc440)
>> [ 237.186230] Stack:
>> [ 237.186231] ffff88011287db50 ffff88011873b780 ffff88011287d800
>> ffffffffa021d3c0
>> [ 237.186233] ffff8801181afdd8 ffffffff8116de93 0000000000000001
>> ffff88011287d800
>> [ 237.186235] ffff8801181afe08 ffffffff811583bc ffff88011873b820
>> 0000000000000015
>> [ 237.186238] Call Trace:
>> [ 237.186241] [<ffffffff8116de93>] shrink_dcache_for_umount+0x33/0x60
>> [ 237.186244] [<ffffffff811583bc>] generic_shutdown_super+0x2c/0xf0
>> [ 237.186247] [<ffffffff81158516>] kill_anon_super+0x16/0x30
>> [ 237.186249] [<ffffffff81158557>] kill_litter_super+0x27/0x30
>> [ 237.186265] [<ffffffffa02174a9>] rpc_kill_sb+0x99/0xc0 [sunrpc]
>> [ 237.186268] [<ffffffff8115877c>] deactivate_locked_super+0x3c/0xa0
>> [ 237.186270] [<ffffffff8115938e>] deactivate_super+0x4e/0x70
>> [ 237.186273] [<ffffffff81174416>] mntput_no_expire+0x106/0x160
>> [ 237.186276] [<ffffffff8117543e>] sys_umount+0x6e/0x3b0
>> [ 237.186279] [<ffffffff81583d96>] system_call_fastpath+0x1a/0x1f
>> [ 237.186280] Code: 00 00 48 8b 40 28 4c 8b 08 48 8b 43 30 48 85 c0
>> 74 04 48 8b 50 40 48 89 34 24 48 c7 c7 48 69 7c 81 48 89 de 31 c0 e8
>> 09 62 40 00 <0f> 0b 0f 0b 66 0f 1f 44 00 00 55 48 89 e5 41 54 53 66
>> 66 66 66
>> [ 237.186301] RIP [<ffffffff8116c606>]
>> shrink_dcache_for_umount_subtree+0x1e6/0x1f0
>> [ 237.186303] RSP <ffff8801181afd98>
>> [ 237.186320] ---[ end trace 74629a120592ec3c ]---
>> [ 238.707925] mtrr: no MTRR for f9000000,e00000 found
>> [ 245.162939] imon:send_packet: task interrupted
>> [ 245.218338] nfsd: last server has exited, flushing export cache
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
Best regards,
Stanislav Kinsbursky
^ permalink raw reply [flat|nested] 2+ messages in thread