linux-kernel - Re: [Linux-cachefs] 3.0.3 64-bit Crash running fscache/cachefilesd

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAOH1cHkg_eQRCHxOxiOPgO=-aZfnjts-tLFQFKffk5JjEfqy2w@mail.gmail.com>
Date:	Wed, 12 Oct 2011 11:10:59 -0700
From:	Mark Moseley <moseleymark@...il.com>
To:	David Howells <dhowells@...hat.com>
Cc:	Linux filesystem caching discussion list 
	<linux-cachefs@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: [Linux-cachefs] 3.0.3 64-bit Crash running fscache/cachefilesd

On Wed, Oct 12, 2011 at 3:05 AM, David Howells <dhowells@...hat.com> wrote:
> Mark Moseley <moseleymark@...il.com> wrote:
>
>> > Hmmm...  An object is being added to the object list twice.  It may
>> > indicate a free before an object is finished with.  Do you have slab
>> > debugging turned on?
>>
>> It's compiled in but not currently enabled. Should I do so? And
>> anything I should be looking for in it?
>
> This raises another question: Which of CONFIG_SLAB/SLOB/SLUB are you actually
> using?
>
> David
>

I had SLUB. I have CONFIG_SLUB_DEBUG but not CONFIG_SLUB_DEBUG_ON.

I've recompiled with SLAB and CONFIG_DEBUG_SLAB enabled though:

CONFIG_SLAB=y
CONFIG_SLABINFO=y
CONFIG_DEBUG_SLAB=y
CONFIG_DEBUG_SLAB_LEAK=y

Any flags anywhere that I need to set?

I've also cleared out the fscache cache. When I initially booted into
the 3.1.0-rc8 with SLAB, I was already at the cachefilesd.conf limit,
space-wise, and it would crash almost immediately (basically as soon
as load balancers started directing traffic to it again). I didn't
have console, so I don't have the full traceback for these ones, but a
little of the top actually made it to the logs:


[  117.654487] FS-Cache: Cache "mycache" added (type cachefiles)
[  117.654494] CacheFiles: File cache on sdb6 registered
[  184.943653] ------------[ cut here ]------------
[  184.952933] kernel BUG at fs/fscache/object-list.c:83!
[  184.960676] invalid opcode: 0000 [#1] SMP
[  184.960676] CPU 1
[  184.960676] Modules linked in: xfs ioatdma dca loop joydev evdev
dcdbas psmouse serio_raw pcspkr i5000_edac edac_core i5k_amb shpchp pc
i_hotplug sg sr_mod cdrom ehci_hcd uhci_hcd sd_mod crc_t10dif [last
unloaded: scsi_wait_scan]
[  184.960676]
[  184.960676] Pid: 638, comm: kworker/u:2 Not tainted 3.1.0-rc8 #1
Dell Inc. PowerEdge 1950/0DT097
[  184.960676] RIP: 0010:[<ffffffff8119d15c>]  [<ffffffff8119d15c>]
fscache_object_destroy+0x4c/0x50
[  184.960676] RSP: 0000:ffff880224cebd90  EFLAGS: 00010246
[  184.960676] RAX: 0000000000000011 RBX: ffff88021920a5e8 RCX: ffff8801d6b92a38
[  184.960676] RDX: 0000000000000010 RSI: 000000000000006b RDI: ffffffff81c94b10

This was the 2nd one, after I'd panicked it starting up cachefilesd a
couple of minutes earlier (which was also "kernel BUG at
fs/fscache/object-list.c:83"), so I figured I should clear the cache.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/