Re: 2.5.66-mm2

[Maciej Soltysiak]

> > remember my post about the machine locking up for a few seconds?
>
> Could you try 2.5.66-mm3?  It has a CPU scheduler fix which might well help here.
It's still there.
Using the following script to get vmstat output...
#!/bin/sh
rm vmlog
while (true);
do
        echo "-----" >> vmlog
        date >> vmlog
        for i in `seq 1 5`; do
                vmstat >> vmlog
                sleep 1;
        done;
done

I managed to get the following. The lockup occured between 9:53:00, and
9:53:20.

pią kwi  4 09:52:58 CEST 2003
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 2  1  65724   4676   7256  48936    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 2  2  65724   4636   8184  49048    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 2  2  65724   4636   8184  49048    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 2  2  65716   5452   7456  49028    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 2  2  65716   5452   7456  49028    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 7 10  65716   4148   7476  49040    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 6 10  65716   4148   7476  49040    8    6    42    44  224   472 98  1  0  0
-----
pią kwi  4 09:53:20 CEST 2003
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 5  6  65736   4532   7344  48804    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 4  6  65736   4532   7348  48804    8    6    42    44  224   472 98  1  0  0
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
-----
pią kwi  4 09:53:21 CEST 2003

Running apps: [debian] xmms + opera + setiathome + x-terminal-emulator

It just stopped responding suddenly for 20 seconds and went on.
I think i need some other method of measuring what is going on.

Notice, that there are only 7 results between these 20 seconds, and there
should be 20. Earlier in the logs i also get these irregularities, the
script can not manage to get it on time, or the script is getting skewed
by some other factors i am unaware of.

Regards,
Maciej