Hi
Searching through the archives I found this post on Tue, Sep 12, 2000 at
09:41:13PM +0200 from Octave Klaba
> Hello,
> On a high load server, kernel has some errors:
>
> VM: do_try_to_free_pages failed for httpd...
> VM: do_try_to_free_pages failed for httpd...
> eth0: Too much work at interrupt, status=0x4050.
> eth0: Too much work at interrupt, status=0x4050.
>
> is there somewhere the new version of driver for eepro100
> to make a test ?
I'm wondering if anyone has found a solution for this: our mailserver is
exhibiting the same error message (although no mention of the eth0
interface, we do also use the eepro100):
Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
sendmail
...
Oct 31 12:12:14 mail-client last message repeated 2 times
Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
in.pop3d
...
Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
klogd...
Oct 31 12:12:23 mail-client last message repeated 14 times
Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
sendmail
...
Oct 31 12:12:23 mail-client last message repeated 15 times
Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
syslogd.
..
Oct 31 12:12:23 mail-client last message repeated 11 times
Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
in.pop3d
...
Oct 31 12:12:23 mail-client last message repeated 15 times
Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
sendmail
...
Oct 31 12:12:24 mail-client last message repeated 30 times
Oct 31 12:12:24 mail-client kernel: VM: do_try_to_free_pages failed for
kupdate.
..
Oct 31 12:12:24 mail-client kernel: VM: do_try_to_free_pages failed for
sendmail
...
Oct 31 12:12:25 mail-client last message repeated 42 times
Oct 31 12:12:25 mail-client kernel: VM: do_try_to_free_pages failed for
kupdate.
..
Oct 31 12:12:25 mail-client kernel: VM: do_try_to_free_pages failed for
sendmail
...
Oct 31 12:12:26 mail-client last message repeated 60 times
Oct 31 12:12:26 mail-client kernel: VM: do_try_to_free_pages failed for
kupdate.
... etc etc
To agree with Octave, this only appears to happen under load - over weekends
(our quiet periods) the syslog is nearly empty. In extremis it has been
necessary to reboot the machine by kicking the power button.
Any thoughts appreciated.
Running Redhat 6.2 (not my choice) on a single-processor PIII-650 with 128MB
RAM and 256MB swap. Load average is usually around the 0.3 mark. Kernel
(2.2.17) was compiled from source.
I'm not subscribed to the list, but I guess this is the only way to find an
answer - can you please copy me in on any replies. Thanks :0
Geoff
> > VM: do_try_to_free_pages failed for httpd...
> > VM: do_try_to_free_pages failed for httpd...
These if they are odd ones and the box continues are fine, if you get masses
of them then probably not
> (our quiet periods) the syslog is nearly empty. In extremis it has been
> necessary to reboot the machine by kicking the power button.
Are you using software raid ?
Alan
"Alan Cox" <[email protected]> writes:
[about what I wrote]
> > > VM: do_try_to_free_pages failed for httpd...
> > > VM: do_try_to_free_pages failed for httpd...
>
> These if they are odd ones and the box continues are fine, if you get
masses
> of them then probably not
What's it actually doing when this happens? Would it help to allocate more
VM?
> > (our quiet periods) the syslog is nearly empty. In extremis it has been
> > necessary to reboot the machine by kicking the power button.
>
> Are you using software raid ?
No. Should have said it's a Symbios 53c896 SCSI onboard, on an Intel GX
board.
Incidentally I stupidly compiled in support for the onboard sound, which is
the es1371 (rev 8) - don't know if this has any bearing but I'll be taking
it out anyway!
Geoff
On Tue, 31 Oct 2000, Geoff Winkless wrote:
> Hi
>
> Searching through the archives I found this post on Tue, Sep 12, 2000 at
> 09:41:13PM +0200 from Octave Klaba
>
> > Hello,
> > On a high load server, kernel has some errors:
> >
> > VM: do_try_to_free_pages failed for httpd...
> > VM: do_try_to_free_pages failed for httpd...
> > eth0: Too much work at interrupt, status=0x4050.
> > eth0: Too much work at interrupt, status=0x4050.
> >
> > is there somewhere the new version of driver for eepro100
> > to make a test ?
>
> I'm wondering if anyone has found a solution for this: our mailserver is
> exhibiting the same error message (although no mention of the eth0
> interface, we do also use the eepro100):
Apply the following patch:
ftp://ftp.kernel.org/pub/linux/kernel/people/andrea/patches/v2.2//2.2.18pre17/VM-global-2.2.18pre17-7.bz
On Tue, 31 Oct 2000, Geoff Winkless wrote:
> "Alan Cox" <[email protected]> writes:
> [about what I wrote]
> > > > VM: do_try_to_free_pages failed for httpd...
> > > > VM: do_try_to_free_pages failed for httpd...
> >
> > These if they are odd ones and the box continues are fine, if you get
> masses
> > of them then probably not
>
> What's it actually doing when this happens? Would it help to allocate more
> VM?
This means try_to_free_pages() function (obviously this function tries to
free pages :)) failed to free any page.
The process will probably free pages (or found free pages around) on the
next run(s) of try_to_free_pages.
I get a similar message with my 2.2.16 kernel:
Oct 23 15:02:12 cartman kernel: VM: do_try_to_free_pages failed for
kswapd...
Oct 23 15:02:12 cartman kernel: VM: do_try_to_free_pages failed for
klogd...
Oct 23 15:02:13 cartman kernel: VM: do_try_to_free_pages failed for
nmbd...
Oct 23 15:02:13 cartman kernel: VM: do_try_to_free_pages failed for
httpd...
Oct 24 14:24:23 cartman kernel: VM: do_try_to_free_pages failed for X...
Oct 24 14:24:23 cartman kernel: VM: do_try_to_free_pages failed for
httpd...
Oct 24 14:24:23 cartman kernel: VM: do_try_to_free_pages failed for
netscape-commun...
Oct 24 14:24:24 cartman kernel: VM: do_try_to_free_pages failed for
mixer_applet...
Oct 24 14:24:24 cartman kernel: VM: do_try_to_free_pages failed for
multiload_apple...
Oct 24 14:24:24 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 24 15:55:43 cartman kernel: VM: do_try_to_free_pages failed for
vmware...
Oct 24 15:55:43 cartman kernel: VM: do_try_to_free_pages failed for
geyes_applet...
Oct 24 17:33:58 cartman kernel: VM: do_try_to_free_pages failed for
deskguide_apple...
Oct 24 17:33:58 cartman kernel: VM: do_try_to_free_pages failed for
xscreensaver...
Oct 24 17:33:58 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 24 17:33:58 cartman kernel: VM: do_try_to_free_pages failed for
mixer_applet...
Oct 24 17:33:58 cartman kernel: VM: do_try_to_free_pages failed for
gnomeicu...
Oct 24 17:33:59 cartman kernel: VM: do_try_to_free_pages failed for
httpd...
Oct 25 15:55:20 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 27 12:37:24 cartman kernel: VM: do_try_to_free_pages failed for
xmms...
Oct 27 12:37:24 cartman kernel: VM: do_try_to_free_pages failed for
gnomeicu...
Oct 27 12:37:24 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 27 12:37:24 cartman kernel: VM: do_try_to_free_pages failed for
xmms...
Oct 27 14:29:28 cartman kernel: VM: do_try_to_free_pages failed for X...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
mixer_applet...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
mini_commander_...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
xscreensaver...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
xmms...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
xmms...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
httpd...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
gpm...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
identd...
Oct 27 14:29:30 cartman kernel: VM: do_try_to_free_pages failed for
init...
Oct 27 14:29:32 cartman kernel: VM: do_try_to_free_pages failed for
mozilla-bin...
Oct 27 14:29:32 cartman kernel: VM: do_try_to_free_pages failed for
xntpd...
Oct 27 17:29:56 cartman kernel: VM: do_try_to_free_pages failed for
xscreensaver...
... notice the "kswapd" and "klogd" at the top.
-M
Geoff Winkless wrote:
>
> Hi
>
> Searching through the archives I found this post on Tue, Sep 12, 2000 at
> 09:41:13PM +0200 from Octave Klaba
>
> > Hello,
> > On a high load server, kernel has some errors:
> >
> > VM: do_try_to_free_pages failed for httpd...
> > VM: do_try_to_free_pages failed for httpd...
> > eth0: Too much work at interrupt, status=0x4050.
> > eth0: Too much work at interrupt, status=0x4050.
> >
> > is there somewhere the new version of driver for eepro100
> > to make a test ?
>
> I'm wondering if anyone has found a solution for this: our mailserver is
> exhibiting the same error message (although no mention of the eth0
> interface, we do also use the eepro100):
>
> Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:14 mail-client last message repeated 2 times
> Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
> in.pop3d
> ...
> Oct 31 12:12:14 mail-client kernel: VM: do_try_to_free_pages failed for
> klogd...
>
> Oct 31 12:12:23 mail-client last message repeated 14 times
> Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:23 mail-client last message repeated 15 times
> Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
> syslogd.
> ..
> Oct 31 12:12:23 mail-client last message repeated 11 times
> Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
> in.pop3d
> ...
> Oct 31 12:12:23 mail-client last message repeated 15 times
> Oct 31 12:12:23 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:24 mail-client last message repeated 30 times
> Oct 31 12:12:24 mail-client kernel: VM: do_try_to_free_pages failed for
> kupdate.
> ..
> Oct 31 12:12:24 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:25 mail-client last message repeated 42 times
> Oct 31 12:12:25 mail-client kernel: VM: do_try_to_free_pages failed for
> kupdate.
> ..
> Oct 31 12:12:25 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:26 mail-client last message repeated 60 times
> Oct 31 12:12:26 mail-client kernel: VM: do_try_to_free_pages failed for
> kupdate.
>
> ... etc etc
>
> To agree with Octave, this only appears to happen under load - over weekends
> (our quiet periods) the syslog is nearly empty. In extremis it has been
> necessary to reboot the machine by kicking the power button.
>
> Any thoughts appreciated.
>
> Running Redhat 6.2 (not my choice) on a single-processor PIII-650 with 128MB
> RAM and 256MB swap. Load average is usually around the 0.3 mark. Kernel
> (2.2.17) was compiled from source.
>
> I'm not subscribed to the list, but I guess this is the only way to find an
> answer - can you please copy me in on any replies. Thanks :0
>
> Geoff
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> Please read the FAQ at http://www.tux.org/lkml/
> Oct 31 12:12:25 mail-client kernel: VM: do_try_to_free_pages failed for
> sendmail
> ...
> Oct 31 12:12:26 mail-client last message repeated 60 times
> Oct 31 12:12:26 mail-client kernel: VM: do_try_to_free_pages failed for
> kupdate.
>
> To agree with Octave, this only appears to happen under load - over weekends
> (our quiet periods) the syslog is nearly empty. In extremis it has been
> necessary to reboot the machine by kicking the power button.
on our servers it arrived only when the swap is used. for exemple:
there is too many requests on the web server, machine begins to swap,
after 3-4 minutes we have the first VM errors. it is too late.
the solutions: reboot server / unplug cable / cut port 80 on the router
and wait ... after 3-4 minutes server rebegins to work finely. If you do
not touch the swap you have never this problem. another solution is to
put lot of ram ;)
Octave