Re: v1.1.22 apparently sleeping

Graham Maltby wrote:
>
> I am seeing short periods (10-15 secs) when squid just stops responding.
>
> There doesn't appear to be anything failing according to any of the log
> files. The only sign of this is the server jumps from an average cpu load
> 20-40% up to 90-110% for 10 secs, then returns to normal.
>
> There doesn't seem to be a shortage of anything (ie. there are free file
> descriptors, free dns servers, etc.) and there isn't any pattern when it
> happens.
>
> Has anyone else seen this happening?
>
> We are running:
>
> linux v2.0.34 (slackware)
> squid v1.1.22
> Intel P200 - 256M ram - 24 gig cache
>
> Squid is acting as a transparent proxy also.
>
> This has only started occuring since we upgraded. Unfortunately, I
> changed two things at once, both Linux .29 to .34 and Squid .20 to .22, so
> I don't know which has introduced the problem.
>

Just a stupidity test...kill 'update' and restart it with 'update -f 1
-s 5' and see if that affects things. You might be getting huge buffer
flushes.

2.0.34 hmm? Not using the aic7xxx driver at all are you? My second bet
is that you have SCSI-bus contention, or SCSI driver problems. 2.0.34 is
supposed to be toxic with adaptec drivers under some setups. I saw a
notification to this effect. Supposedly new drivers are available.

D

-- 
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GAT d- s++: a C++++$ UL++++B+++S+++C++H++U++V+++$ P+++$ L+++ E-
W+++(--)$ N++ w++$>--- t+ 5++ X+() R+ tv b++++ DI+++ e- h-@ 
------END GEEK CODE BLOCK------