All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mike Galbraith <efault@gmx.de>
To: Andrew Morton <akpm@osdl.org>
Cc: Laurent Riffard <laurent.riffard@free.fr>,
	Kernel development list <linux-kernel@vger.kernel.org>
Subject: Re: 2.6.18-rc3-mm2 - OOM storm
Date: Fri, 11 Aug 2006 08:33:51 +0000	[thread overview]
Message-ID: <1155285231.5841.6.camel@Homer.simpson.net> (raw)
In-Reply-To: <20060810021957.38c82311.akpm@osdl.org>

On Thu, 2006-08-10 at 02:19 -0700, Andrew Morton wrote:
> On Thu, 10 Aug 2006 11:04:36 +0200
> Laurent Riffard <laurent.riffard@free.fr> wrote:
> 
> > Le 06.08.2006 12:08, Andrew Morton a écrit :
> > > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.18-rc3/26.18-rc3-mm2/
> > 
> > Hello,
> > 
> > On my system, a cron runs every day to check the integrity of
> > installed RPMS, it runs "rpm -v" on each package, which computes
> > MD5 hash for each installed file and compares this result, the file 
> > size and modification time with values stored in RPM database.
> > 
> > This is the workload. Since 2.6.18-rc3-mm2, this processus eats 
> > all the memory and triggers OOM.
> > 
> > On my system, "free -t" output normally looks like this ("cached" value 
> > is about half of RAM):
> > # free -t 
> >              total       used       free     shared    buffers     cached
> > Mem:        515032     508512       6520          0      22992     256032
> > -/+ buffers/cache:     229488     285544
> > Swap:      1116428        324    1116104
> > Total:     1631460     508836    1122624
> > 
> > After the rpm database check, "free -t" says:
> >              total       used       free     shared    buffers     cached
> > Mem:        515032     507124       7908          0       8132     398296
> > -/+ buffers/cache:     100696     414336
> > Swap:      1116428      34896    1081532
> > Total:     1631460     542020    1089440
> > 
> > And the value of "cached" won't decrease.
> > 
> 
> Yes, I was just trying to reproduce this.  No luck so far.  Will try your
> .config tomorrow.
> 
> It would be interesting to try disabling CONFIG_ADAPTIVE_READAHEAD -
> perhaps that got broken.

I get no oom-killer action, but as soon as memory gets tight, I get
something even more effective.  rpm -qaV reliably emits the below.

kernel BUG at mm/vmscan.c:383!
invalid opcode: 0000 [#1]
4K_STACKS PREEMPT SMP 
last sysfs file: /devices/pci0000:00/0000:00:01.0/0000:01:00.0/resource
Modules linked in: xt_pkttype ipt_LOG xt_limit snd_pcm_oss snd_mixer_oss snd_seq_midi snd_seq_midi_event eeprom snd_seq edd ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables ip6table_filter ip6_tables x_tables nls_iso8859_1 nls_cp437 nls_utf8 saa7134_dvb mt352 video_buf_dvb nxt200x tda1004x tuner snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer sd_mod saa7134 bt878 i2c_i801 snd_page_alloc prism54 ir_kbd_i2c bttv video_buf ir_common ohci1394 snd_mpu401 snd_mpu401_uart btcx_risc tveeprom ieee1394 snd_rawmidi snd_seq_device snd soundcore
CPU:    1
EIP:    0060:[<c105a166>]    Not tainted VLI
EFLAGS: 00210203   (2.6.18-rc3-mm2-smp #162) 
EIP is at remove_mapping+0xa3/0xbf
eax: 80008009   ebx: c1e48200   ecx: c14ad9c0   edx: c1e48200
esi: c1e48200   edi: c14ad9c0   ebp: dffb7e14   esp: dffb7e08
ds: 007b   es: 007b   ss: 0068
Process kswapd0 (pid: 196, ti=dffb7000 task=dffb9a90 task.ti=dffb7000)
Stack: c1e48200 c1e48218 c14ad9c0 dffb7f28 c105a818 dffb7f18 00000000 dffb7f08 
       dffb7f10 c14ad680 c14ad690 dffb7f84 c14ad100 00000020 00000000 00000000 
       00000020 00000000 00000000 c14ad9c0 00000000 00000020 00000000 00000001 
Call Trace:
 [<c105a818>] shrink_inactive_list+0x696/0x8dc
 [<c105aaf0>] shrink_zone+0x92/0xe5
 [<c105b125>] kswapd+0x300/0x40e
 [<c10361d6>] kthread+0xe4/0xe8
 [<c1001005>] kernel_thread_helper+0x5/0xb
DWARF2 unwinder stuck at kernel_thread_helper+0x5/0xb
Leftover inexact backtrace:
 [<c1003f83>] show_stack_log_lvl+0xa6/0xcb
 [<c1004180>] show_registers+0x1d8/0x286
 [<c100437f>] die+0x151/0x333
 [<c10045d9>] do_trap+0x78/0xa3
 [<c1004f16>] do_invalid_op+0x97/0xa1
 [<c13e0369>] error_code+0x39/0x40
 [<c105a818>] shrink_inactive_list+0x696/0x8dc
 [<c105aaf0>] shrink_zone+0x92/0xe5
 [<c105b125>] kswapd+0x300/0x40e
 [<c10361d6>] kthread+0xe4/0xe8
 [<c1001005>] kernel_thread_helper+0x5/0xb
Code: f0 e8 46 88 ff ff 89 f8 e8 ba 5d 38 00 f0 ff 4e 04 b8 01 00 00 00 5b 5e 5f 5d c3 89 f8 e8 a5 5d 38 00 31 c0 eb d4 8b 56 0c eb 8d <0f> 0b 7f 01 6f 75 42 c1 89 f6 e9 6b ff ff ff 0f 0b 7e 01 6f 75 
EIP: [<c105a166>] remove_mapping+0xa3/0xbf SS:ESP 0068:dffb7e08
 <0>------------[ cut here ]------------
kernel BUG at mm/vmscan.c:383!
invalid opcode: 0000 [#2]
4K_STACKS PREEMPT SMP 
last sysfs file: /devices/pci0000:00/0000:00:01.0/0000:01:00.0/resource
Modules linked in: xt_pkttype ipt_LOG xt_limit snd_pcm_oss snd_mixer_oss snd_seq_midi snd_seq_midi_event eeprom snd_seq edd ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables ip6table_filter ip6_tables x_tables nls_iso8859_1 nls_cp437 nls_utf8 saa7134_dvb mt352 video_buf_dvb nxt200x tda1004x tuner snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer sd_mod saa7134 bt878 i2c_i801 snd_page_alloc prism54 ir_kbd_i2c bttv video_buf ir_common ohci1394 snd_mpu401 snd_mpu401_uart btcx_risc tveeprom ieee1394 snd_rawmidi snd_seq_device snd soundcore
CPU:    0
EIP:    0060:[<c105a166>]    Not tainted VLI
EFLAGS: 00210203   (2.6.18-rc3-mm2-smp #162) 
EIP is at remove_mapping+0xa3/0xbf
eax: 80008009   ebx: c1e784a0   ecx: c14ad9c0   edx: c1e784a0
esi: c1e784a0   edi: c14ad9c0   ebp: dfda3ba0   esp: dfda3b94
ds: 007b   es: 007b   ss: 0068
Process rpm (pid: 6150, ti=dfda3000 task=dffca560 task.ti=dfda3000)
Stack: c1e784a0 c1e784b8 c14ad9c0 dfda3cb4 c105a818 dfda3ca4 00000008 dfda3c94 
       dfda3c9c c14ad680 c14ad690 dfda3cf4 c14ad100 00000020 00000000 00000000 
       00000020 00000000 00000000 c14ad9c0 00000000 00000020 00000000 00000001 
Call Trace:
 [<c105a818>] shrink_inactive_list+0x696/0x8dc
 [<c105aaf0>] shrink_zone+0x92/0xe5
 [<c105b68b>] try_to_free_pages+0x157/0x254
 [<c1055c9b>] __alloc_pages+0x155/0x2b4
 [<c1057595>] __do_page_cache_readahead+0x120/0x2a3
 [<c1057806>] ra_dispatch+0xee/0x100
 [<c1057d83>] page_cache_readahead_adaptive+0x3f4/0xb77
 [<c105349e>] filemap_nopage+0x41d/0x4ad
 [<c105e80d>] __handle_mm_fault+0x12e/0x8fb
 [<c101966a>] do_page_fault+0xdc/0x51f
 [<c13e0369>] error_code+0x39/0x40
 [<b7bc89cf>] 0xb7bc89cf
 [<c1003f83>] show_stack_log_lvl+0xa6/0xcb
 [<c1004180>] show_registers+0x1d8/0x286
 [<c100437f>] die+0x151/0x333
 [<c10045d9>] do_trap+0x78/0xa3
 [<c1004f16>] do_invalid_op+0x97/0xa1
 [<c13e0369>] error_code+0x39/0x40
 [<c105a818>] shrink_inactive_list+0x696/0x8dc
 [<c105aaf0>] shrink_zone+0x92/0xe5
 [<c105b68b>] try_to_free_pages+0x157/0x254
 [<c1055c9b>] __alloc_pages+0x155/0x2b4
 [<c1057595>] __do_page_cache_readahead+0x120/0x2a3
 [<c1057806>] ra_dispatch+0xee/0x100
 [<c1057d83>] page_cache_readahead_adaptive+0x3f4/0xb77
 [<c105349e>] filemap_nopage+0x41d/0x4ad
 [<c105e80d>] __handle_mm_fault+0x12e/0x8fb
 [<c101966a>] do_page_fault+0xdc/0x51f
 [<c13e0369>] error_code+0x39/0x40
Code: f0 e8 46 88 ff ff 89 f8 e8 ba 5d 38 00 f0 ff 4e 04 b8 01 00 00 00 5b 5e 5f 5d c3 89 f8 e8 a5 5d 38 00 31 c0 eb d4 8b 56 0c eb 8d <0f> 0b 7f 01 6f 75 42 c1 89 f6 e9 6b ff ff ff 0f 0b 7e 01 6f 75 
EIP: [<c105a166>] remove_mapping+0xa3/0xbf SS:ESP 0068:dfda3b94
 



  parent reply	other threads:[~2006-08-11  6:26 UTC|newest]

Thread overview: 131+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-08-06 10:08 2.6.18-rc3-mm2 Andrew Morton
2006-08-06 11:09 ` 2.6.18-rc3-mm2 Michal Piotrowski
2006-08-07  9:52   ` 2.6.18-rc3-mm2 Balbir Singh
2006-08-07 12:16     ` 2.6.18-rc3-mm2 Michal Piotrowski
2006-08-07 14:05       ` 2.6.18-rc3-mm2 Balbir Singh
2006-08-06 13:33 ` 2.6.18-rc3-mm2 Mattia Dongili
2006-08-06 14:55   ` 2.6.18-rc3-mm2 [BUG at mm/vmscan.c:383!] Hugh Dickins
2006-08-06 17:02     ` Mattia Dongili
2006-08-06 14:11 ` 2.6.18-rc3-mm2 Reuben Farrelly
     [not found] ` <b637ec0b0608060848k22af58cbo6f13cee19498c2d2@mail.gmail.com>
2006-08-06 19:09   ` 2.6.18-rc3-mm2 Andrew Morton
2006-08-07  2:18     ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-07 18:47       ` 2.6.18-rc3-mm2 Fabio Comolli
2006-08-07 19:00         ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-08 14:41           ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-08 17:42             ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-08 18:16               ` 2.6.18-rc3-mm2 Fabio Comolli
2006-08-08 18:24                 ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-08 18:36                   ` 2.6.18-rc3-mm2 Fabio Comolli
2006-08-09  3:47                     ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-09  7:11                       ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-09 19:47                       ` 2.6.18-rc3-mm2 Fabio Comolli
2006-08-09 20:13                         ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-08 20:32               ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-08 18:14           ` 2.6.18-rc3-mm2 Fabio Comolli
2006-08-06 22:42 ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-06 22:54   ` 2.6.18-rc3-mm2 Andrew Morton
2006-08-07  9:15     ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-07 20:34       ` 2.6.18-rc3-mm2 Rafael J. Wysocki
2006-08-07 20:55         ` 2.6.18-rc3-mm2 Andrew Morton
2006-08-08  5:21           ` 2.6.18-rc3-mm2 Jens Axboe
2006-08-07  2:18   ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-07  2:20     ` 2.6.18-rc3-mm2 Dmitry Torokhov
2006-08-07  2:07 ` 2.6.18-rc3-mm2 Grant Coady
2006-08-07  9:28 ` swsusp regression [Was: 2.6.18-rc3-mm2] Jiri Slaby
2006-08-07 16:23   ` Jason Lunz
2006-08-07 20:47     ` Rafael J. Wysocki
2006-08-07 20:47       ` Rafael J. Wysocki
2006-08-08  8:41       ` Jens Axboe
2006-08-08  9:49         ` Jiri Slaby
2006-08-08 10:43           ` Jens Axboe
2006-08-08 10:08       ` Jiri Slaby
2006-08-08 10:43         ` Jens Axboe
2006-08-08 10:43           ` Jens Axboe
2006-08-08 10:59           ` Rafael J. Wysocki
2006-08-08 11:04             ` Jens Axboe
2006-08-08 11:07               ` Jens Axboe
2006-08-08 11:16                 ` Rafael J. Wysocki
2006-08-08 11:16                   ` Rafael J. Wysocki
2006-08-08 11:19                   ` Jens Axboe
2006-08-08 13:50                     ` Rafael J. Wysocki
2006-08-08 13:50                       ` Rafael J. Wysocki
2006-08-08 14:06                       ` Jens Axboe
2006-08-08 16:41                         ` Jiri Slaby
2006-08-08 17:53                           ` Jens Axboe
2006-08-07 21:09     ` Jiri Slaby
2006-08-07 13:40 ` x86_64 command line truncated Andy Whitcroft
2006-08-07 14:05   ` Andi Kleen
2006-08-07 14:37     ` x86_64 command line truncated II Andi Kleen
2006-08-07 14:42       ` Andy Whitcroft
2006-08-07 14:46         ` Andi Kleen
2006-08-07 15:04           ` Andy Whitcroft
2006-08-07 15:12             ` [PATCH] x86_64 dirty fix to restore dual command line store Andy Whitcroft
2006-08-07 21:47               ` Keith Mannthey
2006-08-07 21:59                 ` Keith Mannthey
2006-08-07 14:38     ` x86_64 command line truncated Andy Whitcroft
2006-08-07 15:15       ` Andrew Morton
2006-08-07 15:58         ` Andi Kleen
2006-08-07 15:49 ` [-mm patch] make arch/i386/kernel/acpi/boot.c:acpi_force static Adrian Bunk
2006-08-07 16:07   ` Andi Kleen
2006-08-07 15:49 ` [-mm patch] make arch/i386/kernel/apic.c:enable_local_apic static Adrian Bunk
2006-08-07 15:49 ` [-mm patch] net/: make code static Adrian Bunk
2006-08-08  4:51   ` David Miller
2006-08-07 15:50 ` [-mm patch] drivers/crypto/geode-aes.c: cleanups Adrian Bunk
2006-08-07 19:38 ` resume from S3 regression [Was: 2.6.18-rc3-mm2] Mattia Dongili
2006-08-07 19:38   ` Mattia Dongili
2006-08-07 20:02   ` Andrew Morton
2006-08-07 20:57     ` Mattia Dongili
2006-08-07 22:09       ` Mattia Dongili
2006-08-07 21:04 ` [RFC: -mm patch] bcm43xx_main.c: remove 3 functions Adrian Bunk
2006-08-08 18:32   ` Michael Buesch
2006-08-08 19:42     ` Adrian Bunk
2006-08-09  4:47       ` Michael Buesch
2006-08-08 22:14     ` Jeff Garzik
2006-08-08 14:39 ` 2.6.18-rc3-mm2: reiserfs problem? Rafael J. Wysocki
2006-08-08 15:12   ` Andrew Morton
     [not found]   ` <20060804192540.17098.39244.stgit@warthog.cambridge.redhat.com>
2006-08-08 17:23     ` [PATCH] ReiserFS: Make sure all dentries refs are released before calling kill_block_super() David Howells
2006-08-08 23:16       ` Rafael J. Wysocki
2006-08-09 10:14         ` David Howells
2006-08-09 10:23           ` Rafael J. Wysocki
2006-08-09 11:00         ` David Howells
2006-08-09 13:43     ` [PATCH] ReiserFS: Make sure all dentries refs are released before calling kill_block_super() [try #2] David Howells
2006-08-09 21:56       ` Rafael J. Wysocki
2006-08-10 10:16         ` David Howells
2006-08-09 19:06 ` 2.6.18-rc3-mm2 - ext3 locking issue? Valdis.Kletnieks
2006-08-09 20:01   ` Andrew Morton
2006-08-09 20:43     ` Valdis.Kletnieks
2006-08-10  3:32       ` Valdis.Kletnieks
2006-08-10 11:40         ` Jiri Slaby
2006-08-10 15:27           ` Andrew Morton
2006-08-10 17:33             ` Mattia Dongili
2006-08-10 17:43               ` Jiri Slaby
2006-08-10 17:44               ` Valdis.Kletnieks
2006-08-11  6:17                 ` Andrew Morton
2006-08-11  6:55                   ` Valdis.Kletnieks
2006-08-11 22:39           ` Laurent Riffard
2006-08-15 23:38   ` Valdis.Kletnieks
2006-08-10  9:04 ` 2.6.18-rc3-mm2 - OOM storm Laurent Riffard
2006-08-10  9:19   ` Andrew Morton
2006-08-10 23:20     ` Laurent Riffard
2006-08-11 12:31       ` Laurent Riffard
2006-08-11 21:50         ` Mike Galbraith
2006-08-11  8:33     ` Mike Galbraith [this message]
2006-08-11  6:55       ` Andrew Morton
2006-08-11  9:37         ` Mike Galbraith
2006-08-12 15:07     ` [patch] " Mike Galbraith
2006-08-12 21:26       ` Laurent Riffard
2006-08-10 12:13 ` [patch] Use rwsems instead of custom locking scheme in net/socket.c and net/dccp/ccid.c Frederik Deweerdt
2006-08-10 12:57   ` David Miller
2006-08-10 13:19     ` Frederik Deweerdt
2006-08-10 13:43 ` 2.6.18-rc3-mm2 [oops: shrink_dcache_for_umount_subtree ?] Reuben Farrelly
2006-08-10 15:38   ` Andrew Morton
2006-08-10 17:38 ` 2.6.18-rc3-mm2 - IPV6_MULTIPLE_TABLES borked Valdis.Kletnieks
2006-08-10 20:02   ` Patrick McHardy
2006-08-10 21:44     ` Valdis.Kletnieks
2006-08-11  2:15 ` 2.6.18-rc3-mm2 - BUG in rt6_lookup() from ipv6_del_addr() Valdis.Kletnieks
2006-08-11  4:20   ` David Miller
2006-08-11 18:11 ` 2.6.18-rc3-mm2 Mark Haverkamp
2006-08-11 18:36   ` 2.6.18-rc3-mm2 Andrew Morton
2006-08-11 20:31     ` 2.6.18-rc3-mm2 Mark Haverkamp
2006-08-11 22:58       ` 2.6.18-rc3-mm2 Andrew Morton
2006-08-23 17:02         ` 2.6.18-rc3-mm2 Mark Haverkamp

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1155285231.5841.6.camel@Homer.simpson.net \
    --to=efault@gmx.de \
    --cc=akpm@osdl.org \
    --cc=laurent.riffard@free.fr \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.