public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: "Haar János" <djani22@netcenter.hu>
To: "Chris Wedgwood" <cw@f00f.org>
Cc: <linux-kernel@vger.kernel.org>
Subject: Re: swapper: page allocation failure.
Date: Sun, 21 May 2006 10:37:00 +0200	[thread overview]
Message-ID: <010801c67cb1$bc13fd00$1800a8c0@dcccs> (raw)
In-Reply-To: 20060521081621.GA1151@taniwha.stupidest.org


----- Original Message ----- 
From: "Chris Wedgwood" <cw@f00f.org>
To: "Haar J?nos" <djani22@netcenter.hu>
Cc: <linux-kernel@vger.kernel.org>
Sent: Sunday, May 21, 2006 10:16 AM
Subject: Re: swapper: page allocation failure.


> On Sun, May 21, 2006 at 10:10:13AM +0200, Haar J?nos wrote:
>
> > This server have 2GB ram, and ~1.1G always free!
> > Anybody have an idea?
>
> you ran out of lowmem?
>
> what kernel is this and how do you trigger it?

I have 4 disk nodes that acts as nbd-server.

The only processes:

  PID TTY      STAT   TIME COMMAND
    1 ?        S      0:12 init [3]
    2 ?        SW     0:00 [migration/0]
    3 ?        SWN    0:19 [ksoftirqd/0]
    4 ?        SW     0:00 [migration/1]
    5 ?        SWN    0:03 [ksoftirqd/1]
    6 ?        SW<    0:00 [events/0]
    7 ?        SW<    0:00 [events/1]
    8 ?        SW<    0:00 [khelper]
    9 ?        SW<    0:00 [kthread]
   12 ?        SW<    0:07  \_ [kblockd/0]
   13 ?        SW<    0:02  \_ [kblockd/1]
   14 ?        SW<    0:00  \_ [kacpid]
  101 ?        SW<    0:00  \_ [khubd]
  103 ?        SW<    0:00  \_ [kseriod]
  247 ?        SW    12:09  \_ [pdflush]
  248 ?        SW    10:04  \_ [pdflush]
  250 ?        SW<    0:00  \_ [aio/0]
  251 ?        SW<    0:00  \_ [aio/1]
  252 ?        SW<    0:00  \_ [xfslogd/0]
  253 ?        SW<    0:00  \_ [xfslogd/1]
  254 ?        SW<    0:00  \_ [xfsdatad/0]
  255 ?        SW<    0:00  \_ [xfsdatad/1]
  921 ?        SW<    0:00  \_ [ata/0]
  922 ?        SW<    0:00  \_ [ata/1]
  926 ?        SW<    0:00  \_ [scsi_eh_0]
  927 ?        SW<    0:00  \_ [scsi_eh_1]
  935 ?        SW<    0:00  \_ [scsi_eh_2]
  936 ?        SW<    0:00  \_ [scsi_eh_3]
  937 ?        SW<    0:00  \_ [scsi_eh_4]
  938 ?        SW<    0:00  \_ [scsi_eh_5]
  951 ?        SW<    0:00  \_ [scsi_eh_6]
  952 ?        SW<    0:00  \_ [scsi_eh_7]
 1045 ?        SW<    0:00  \_ [exec-osm/0]
 1047 ?        SW<    0:00  \_ [exec-osm/1]
 1088 ?        SW<    0:00  \_ [kcryptd/0]
 1089 ?        SW<    0:00  \_ [kcryptd/1]
 1090 ?        SW<    0:00  \_ [kmpathd/0]
 1091 ?        SW<    0:00  \_ [kmpathd/1]
 1092 ?        SW<    0:00  \_ [kmirrord]
 1093 ?        SW<    0:14  \_ [kedac]
 1111 ?        SW<  219:01  \_ [md0_raid5]
 1112 ?        SW<    0:23  \_ [rpciod/0]
 1113 ?        SW<    0:00  \_ [rpciod/1]
  249 ?        SW    51:30 [kswapd0]
 2532 ?        S      0:05 syslogd -m 0
 2536 ?        S      0:00 klogd -x
 2615 ?        S      0:00 /usr/sbin/sshd
21484 ?        S      0:00  \_ /usr/sbin/sshd
21486 pts/2    S      0:00      \_ -bash
22444 pts/2    R      0:00          \_ ps fax
 2629 ?        S      0:03 crond
 2638 ?        S      0:55 /bin/bwbar --text-file
/mnt/DY_SYSTEM/sysinfo/st4-eth0_o.php --png-file
/mnt/DY_SYSTEM/sysinfo/st4-eth0_o.png eth0 100 -x 300
 2639 ?        S      0:54 /bin/bwbar --text-file
/mnt/DY_SYSTEM/sysinfo/st4-eth0_i.php -i --png-file
/mnt/DY_SYSTEM/sysinfo/st4-eth0_i.png eth0 100 -x 300
 2674 tty2     S      0:00 /sbin/mingetty tty2
 2675 tty3     S      0:00 /sbin/mingetty tty3
 2676 tty4     S      0:00 /sbin/mingetty tty4
 2677 tty5     S      0:00 /sbin/mingetty tty5
 2678 tty6     S      0:00 /sbin/mingetty tty6
 2925 tty1     S      0:00 /sbin/mingetty tty1 --noclear
 5167 ?        SW     0:00 [lockd]
21290 ?        S<     0:00 /usr/local/bin/nbd-server 1230 /dev/md0
21780 ?        S<     0:38  \_ /usr/local/bin/nbd-server 1230 /dev/md0

The kernel :  2.6.17-rc3-git1, but if i have right, the 2.6.16.1 is the same
if i try to swich back. :-)

2 of 4 nodes frequently reboots without any error message.
I have turned on some debug option, and catch this message.

At this point i watch the free -l -s 1, and i can see, the nbd-server (or
kernel) didn't use the highmem for buffer cache, but why?

             total       used       free     shared    buffers     cached
Mem:       2073048     893360    1179688          0     829092      19820
Low:        893464     868352      25112          0          0          0
High:      1179584      25008    1154576          0          0          0
-/+ buffers/cache:      44448    2028600
Swap:            0          0          0

(min_free_kbytes = 16000 now.)

Cheers,
Janos


  reply	other threads:[~2006-05-21  8:37 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-05-21  8:10 swapper: page allocation failure Haar János
2006-05-21  8:16 ` Chris Wedgwood
2006-05-21  8:37   ` Haar János [this message]
2006-05-21  8:47     ` Chris Wedgwood
2006-05-21  9:03       ` Haar János
2006-05-21  9:10         ` Chris Wedgwood
2006-05-21  9:31           ` Haar János
2006-05-21 10:24             ` Chris Wedgwood
2006-05-21 10:42               ` Haar János
2006-05-21 10:26             ` Chris Wedgwood
2006-05-21 10:45               ` Haar János
2006-05-21 12:01               ` Nick Piggin
2006-05-21 13:58                 ` Haar János
2006-05-22  0:09                   ` Nick Piggin
2006-05-22  7:41                     ` Haar János
2006-05-22 11:17                       ` Con Kolivas
2006-05-22 15:08                         ` Haar János
2006-05-22 15:12                           ` Con Kolivas
2006-05-22 21:06                             ` swapper: page allocation failure. - random reboot problem Haar János
2006-05-23  9:17                               ` Nick Piggin
2006-05-23 10:16                                 ` Haar János
2006-05-23 10:24                                   ` Nick Piggin
2006-05-23 10:31                                     ` Haar János
2006-05-23 10:45                                       ` Nick Piggin
2006-05-23 11:01                                         ` Haar János
2006-05-21  9:23       ` swapper: page allocation failure Haar János
2006-05-21 12:03 ` Andrew Morton
2006-05-21 14:01   ` Haar János

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='010801c67cb1$bc13fd00$1800a8c0@dcccs' \
    --to=djani22@netcenter.hu \
    --cc=cw@f00f.org \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox