Hi all,

On a remote machine I'm running Gentoo with the Gentoo Stable kernel.
However, I've been experiencing some problems with it.

First event is that after about 12 days the load went up to around 18-20.
Getting a process listing with 'ps ax' resulted in a hang, as did the
command 'w'. Going through /proc/<pid>/ by hand I was able to find out a
single process id that presumably caused these hangings and high load. It
turned out to be an Apache2 process. Rebooting was also impossible due to
the hanging process, so the only solution turned out to be resetting the
machine.

Second event turned up again after about 12 days uptime. This time the load
went up to a full load of 1.00 (and goes up if the machine has some real
things to do like accepting mail or HTTP requests). I'm also seeing one
zombie being present in the form of kswapd being defunct. The machine is
still responding correctly though.

Third problem is that I've received two errors from Mailman, both times from
the gate_news script run by cron. Perhaps these are related to kswapd being
defunct, I don't know. The errors are "free(): invalid pointer 0x814d3e0"
and "Fatal Python error: Inconsistent interned string state."

I'm hoping somebody has a clue as to why this machine running Gentoo Stable
is proofing to be unstable?

Many thanks in advance,
Kind regards,

Guido

Johnny Doe enlightened us with:
> On a remote machine I'm running Gentoo with the Gentoo Stable
> kernel. However, I've been experiencing some problems with it.

Ok.

> [ snipped process hangs and zombie kswapd ]
>
> Third problem is that I've received two errors from Mailman, both
> times from the gate_news script run by cron. Perhaps these are
> related to kswapd being defunct, I don't know. The errors are
> "free(): invalid pointer 0x814d3e0" and "Fatal Python error:
> Inconsistent interned string state."

This looks like a hardware problem to me. Check your memory using
memtest86. Faulty memory can cause weird errors and is often difficult
to diagnose, but it's really easy to test.

> I'm hoping somebody has a clue as to why this machine running Gentoo
> Stable is proofing to be unstable?

Unstable hardware?

Sybren
--
The problem with the world is stupidity. Not saying there should be a
capital punishment for stupidity, but why don't we just take the
safety labels off of everything and let the problem solve itself?

Johnny Doe wrote:

<defunct swap monitor>

> Third problem is that I've received two errors from Mailman, both times from
> the gate_news script run by cron. Perhaps these are related to kswapd being
> defunct, I don't know. The errors are "free(): invalid pointer 0x814d3e0"
> and "Fatal Python error: Inconsistent interned string state."

Test your memory and your disk(s).

--
J

All your bits are belong to us - again.

> This looks like a hardware problem to me. Check your memory using
> memtest86. Faulty memory can cause weird errors and is often
> difficult to diagnose, but it's really easy to test.

> Test your memory and your disk(s).

Sybren, Jeroen, thanks for your input. I'm going to check the memory ASAP
and I hope that I'll indeed be discovering a bad module. (I already tested
2x 128 MB modules with Memtest86 before they were put into the machine,
however, the orignal 128 MB and 256 MB modules have not yet been tested with
Memtest86, so they might indeed be corrupt).

The discs are brand new and are being watched with SMART, so I think it's
unlikely that it might be the discs. Taking into account that the old
installation (which was running on the same CPU and the original 128 MB and
256 MB memory modules, different mobo, different HDDs) was also having
strange lockups, my guess goes to the memory as well.

Thanks,
Kind regards,

Guido

try using a different version of python.


http://www.google.com/search?hl=de&i...le-Suche&meta=