Re: netbsd-4 vs -current web server performance

[Jeff Rizzo <riz%NetBSD.org@localhost>]

Thor Lancelot Simon wrote:
> On Tue, Sep 02, 2008 at 02:52:33PM -0700, Jeff Rizzo wrote:
>   
>> Jeff Rizzo wrote:
>>     
>>> Looking at both machines, the one running 4 was at its normal load -
>>> between 4 and 5 - but the machine running -current had a load of about
>>> 12, and many of the httpd processes were showing as 'parked' in top, and
>>> there were no idle cpus. Restarting apache on the affected machine seems
>>> to have brought things back to normal - both machines have a load just
>>> under 3, and all CPUs (8 per box) have ~80% idle.
>>>   
>>>       
>> By way of following up to myself, I just had to pull the -current
>> machine out of the rotation for unrelated reasons (forgot to install an
>> important package on it), and the load dropped away *except* for three
>> httpd processes:
>>
>>   PID USERNAME PRI NICE   SIZE   RES STATE      TIME   WCPU    CPU COMMAND
>>  4742 www       42    0    17M  314M parked/2 233:14 99.02% 99.02% httpd
>>  3265 www       39    0    17M  266M parked/0 164:34 99.02% 99.02% httpd
>>  8844 www       40    0    17M  298M parked/0 158:27 99.02% 99.02% httpd
>>     
>
> Parked *and* 100% of a CPU?
>
> That seems just wrong.
>
> Thor
>   

I would tend to agree - but I don't know all that much about the new
states.  This issue is fairly reproducible (within a few hours) - and
also seems to go away (for a few hours) after restarting apache.  I have
a couple weeks to get this fixed, since even at it's worst it's not too
bad, and in any event I have multiple webservers.

+j