Re: [Fwd: Re: [squid-users] Recommended cache_dir config for large system]

From: Adrian Chadd <adrian_at_freebsd.org>
Date: Thu, 17 Jul 2008 10:25:34 +0800

If you're running SATA then it could just be a crappy ATA chipset. you
should see how much time you're spending in iowait and whether that
correlates with your "D" processes.

Adrian

2008/7/17 Richard Hubbell <richard_hubbe11_at_yahoo.com>:
>
>
>
> --- On Wed, 7/16/08, Rhino <rhino_at_machlink.com> wrote:
>
>> From: Rhino <rhino_at_machlink.com>
>> Subject: Re: [Fwd: Re: [squid-users] Recommended cache_dir config for large system]
>> To: richard_hubbe11_at_yahoo.com
>> Cc: squid-users_at_squid-cache.org
>> Date: Wednesday, July 16, 2008, 6:55 PM
>> Richard Hubbell wrote:
>> >
>> > --- On Wed, 7/16/08, Rhino <rhino_at_machlink.com>
>> wrote:
>> >
>> >
>> >> From: Rhino <rhino_at_machlink.com>
>> >> Subject: Re: [Fwd: Re: [squid-users] Recommended
>> cache_dir config for large system]
>> >> To: richard_hubbe11_at_yahoo.com
>> >> Cc: squid-users_at_squid-cache.org
>> >> Date: Wednesday, July 16, 2008, 2:53 PM
>> >> Richard Hubbell wrote:
>> >>
>> >>>> THanks much for the quick response,
>> Henrik.
>> >>>> Filesystem for cache disks currently
>> configured
>> >>>>
>> >> for
>> >>
>> >>>> reiserfs with
>> >>>> notail/noatime opts.
>> >>>> I did not have the fd amounts set, nor
>> >>>>
>> >> ip_local_port_range.
>> >>
>> >>>> My cache_dirs have each disk mounted as
>> partition,
>> >>>>
>> >> i.e.
>> >>
>> >>>> disk1=/squid1
>> >>>> disk2=/squid2; would your suggestion
>> >>>> be then to halve each disk and partition
>> each as
>> >>>>
>> >> cache_dir?
>> >>
>> >>>> (i.e, go
>> >>>> from squid1-4 to squid1-8 across the 4
>> disks)
>> >>>> Also have a 5th disk of equal size that
>> has to be
>> >>>>
>> >> used for
>> >>
>> >>>> OS, just fyi
>> >>>> - so these 4 are totally dedicated to
>> Squid.
>> >>>> thanks again, appreciate your input.
>> >>>> -Ryan
>> >>>>
>> >>> Just curious why reiserfs? I don't think
>> it's
>> >>>
>> >> supported any longer.
>> >>
>> >>>
>> >>>
>> >>>
>> >>>
>> >>>
>> >> size/speed considerations when we set the system
>> up
>> >> originally. It's
>> >> worked well so far.
>> >> cheers
>> >>
>> >
>> > Okay, do you mean better performance with larger
>> number of files?
>> > I'm using ext3 that's why I'm asking.
>> >
>> >
>> >
>> >
>> >
>> >
>> >
>> we went with the reiserfs out of googling the choices and
>> most impressed
>> with what we found on it as opposed to ext3.
>> we're also using sata drives, if that helps; just hoped
>> for the best
>> combination once the server was in production.
>> I'm not suggesting this is in fact the optimum
>> combination - I'm no
>> linux or filesystem expert by any means.
>> Just trying to keep our customers happy and my paycheck
>> coming :)
>> -Ryan
>
> Yep, no problem. We're using ext3 on redhat since that's what's supported. Always good to hear/learn from others' experience.
>
> We've seen some issues with processes spending a litte too much time in "D" state (as reported by ps) a.k.a. uninteruptible sleep. Not clear what is causing that yet. Seems like a deadlock somewhere. Not in squid. We saw it in apache and others.
>
>
>
>
>
Received on Thu Jul 17 2008 - 02:25:37 MDT

This archive was generated by hypermail 2.2.0 : Thu Jul 17 2008 - 12:00:03 MDT