Re: How to tell that my cache is overloaded...

On Thursday 3 July 97, at 20 h 54, the keyboard of
Arjan.deVet@adv.IAEhv.nl (Arjan de Vet) wrote:

> I wrote a timer patch (see http://www.IAEhv.nl/users/devet/squid/) which
> reports a.o. the time Squid spends waiting for select(2) to return:

I highly recommend this patch (unfortunately, it doesn't apply cleanly on
1.1.9 and later, you have to do it by hand). I suggest the patch to be
included in the next release (may be with a configure option).

> 97/07/03 22:40:41| time 30016 idle 27174(6233) diskread 217(77) diskwrite 40(
227) ...
>
> This means that during the last 30.016 seconds Squid was waiting 27.174
> "idle" seconds for select(2) to indicate that new data is available on one
> of the filesdescriptors. This means it was not busy at all :-). The

Here are typical values for the Renater national cache (unlink is always
zero since unlinkd appeared in 1.1.9):

97/07/04 09:33:43| time 30002 idle 21086(16139) diskread 710(287)
diskwrite 508(850) unlink 0(0) maintainSwapSpace 15(30) storeDirClean
0(0) ipcache 33(3) acl 90(473) timeouts 27(30) readhdl 6091(2406)
writehdl 1156(3467) readhdli 405(448) writehdli 82(295) netread 142(1540)
netwrite 285(2405)

Before 1.1.9, unlink used a significant amount of time.
Received on Fri Jul 04 1997 - 00:42:49 MDT