lkml.org 
[lkml]   [1999]   [Feb]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 2.0.36 kswapd gone insane?
On Mon, Feb 22, 1999 at 09:18:19PM -0700, Michael Loftis wrote:
> The system is an i486/66 computer and based on the uptime the amount of
> time kswapd has spent is impossible. Right about 4,637 minutes of uptime.
> The system hasn't been under unusual stress. Earlier today the httpd got
> swapped out and took an irate amount of time to swap in. No errors in
> /var/log/messages Nothing on console, and dmesg leaves no clues except
> for a kludge of lines:
>
> MASQ: failed TCP/UDP checksum from 216.119.135.31!
> MASQ: failed TCP/UDP checksum from 216.119.135.31!
> MASQ: failed TCP/UDP checksum from 216.119.135.31!
> MASQ: failed TCP/UDP checksum from 216.119.135.31!
> MASQ: failed TCP/UDP checksum from 216.119.135.31!
> MASQ: failed TCP/UDP checksum from 216.119.135.31!

> [2107][root@disks:/home/winbox/ck]$ ps x
> PID TTY STAT TIME COMMAND
> 1 ? S 0:04 init
> 2 ? SW 71224874:29 (kflushd)
> 3 ? SW< 0:18 (kswapd)
> 4 ? SW 0:00 (md_thread)
> 5 ? SW 0:00 (md_thread)
> 36 ? S 0:00 /sbin/kerneld
>
> ---SNIP---
> [2108][root@disks:/home/winbox/ck]$ uptime
> 9:09pm up 3 days, 5:17, 4 users, load average: 0.10, 0.10, 0.09
> [2109][root@disks:/home/winbox/ck]$

I have seen this before and so have others. Some old messages from
linux-kernel:

On Tue, Jan 21, 1997 at 03:34:56PM +0200, Petri Kaukasoina wrote:
> Something strange just happened. Reading news became very very slow
> and syslog received these:
>
> Jan 21 15:23:08 elektroni kernel: eth0: Infinite loop in interrupt, status 2011.
> Jan 21 15:23:08 elektroni last message repeated 10 times
> Jan 21 15:24:01 elektroni kernel: eth0: Infinite loop in interrupt, status 2011.
>
> After a minute or so everything seems ok again.
>
> But kflushed has gained a "little" cpu time:
>
> USER PID %CPU %MEM SIZE RSS TTY STAT START TIME COMMAND
> root 2 0.0 0.0 0 0 ? SW Jan 14 71224874:23 (kflushd)
>
> Kernel 2.0.28, 486/66, IDE, ext2, VLB, 16 M RAM, 3c509b. IP-firewall. Xfree.

On Thu, Mar 05, 1998 at 12:48:36AM -0500, Jon Lewis wrote:
> Speaking of kswapd...anyone ever seen kswapd eat LOTS of CPU time?
> Today, or Squid server started acting funny (atually, it just stopped
> processing requests even though it was running and eating fds) and the
> only other odd thing I noticed was:
>
> root 3 0.0 0.0 0 0 ? SW<Mar 3 71224874:27 (kswapd)
>
> The system's only been up 1 day. Is the kernel (2.0.33) confused?

On Sun, Jun 28, 1998 at 04:54:02AM +1200, Spudgun wrote:

> spudgun@potato:~$ uname -a
> Linux potato 2.0.34 #6 Sun Jun 21 19:36:44 NZST 1998 i486 unknown
>
> spudgun@potato:~$ uptime
> 4:39am up 3 days, 3:23, 7 users, load average: 0.27, 0.72, 0.44
>
> spudgun@potato::~$ ps aux
> USER PID %CPU %MEM SIZE RSS TTY STAT START TIME COMMAND
> nobody 803 0.0 5.7 5052 1792 1 S Jun 25 1:00 squid -D -sY
> - --*snip*--
> root 1 0.0 0.2 848 68 ? S Jun 25 0:06 init [5]
> root 2 0.0 0.0 0 0 ? SW Jun 25 71224874:50 (kflushd)
> root 3 0.0 0.0 0 0 ? SW<Jun 25 0:13 (kswapd)
> root 9 0.0 0.1 824 40 ? S Jun 25 0:14 /sbin/update
> root 10 0.0 1.7 1304 536 ? S Jun 25 0:42 /sbin/kerneld
> - --*snip*--

On Tue, Jun 30, 1998 at 01:29:38AM +0300, Meelis Roos wrote:
> I searched the articles of linux-kernel since June 1996 for subjects containing
> kswapd or kflushd and found nothing interesting. I have this little problem
> too, either kflushd or kswapd (don't remember which) shows _very_ much
> used time. 2.0.33pre-someting probably, or stock 2.0.33. Long uptime.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.039 / U:0.064 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site