Re: [squid-users] squid becomes very slow during peak hours

Hi,

  I am running both 2.7-STABLE6 & 3.1.0.8 versions on
more then a few servers. On avg I am pushing about 230+ TPS and at Peak
usage I don't see delays from SQUID unless you come across a content
server that is having issues or isn't cache freindly, DNS issues will
cause problems... but I can't stress enough how fast you should ditch
RAID all together and go with a JBOD and you will see a big improvement
now matter what OS you are running.

Quin

----- Original Message ----
From: goody goody <thinkodd_at_yahoo.com>
To: squid-users_at_squid-cache.org
Cc: Chris Robertson <crobertson_at_gci.net>; balique8061_at_yahoo.com; henrik_at_henriknordstrom.net; Amos jafferies Squid GURU <squid3_at_treenet.co.nz>
Sent: Wednesday, July 1, 2009 7:07:59 AM
Subject: Re: [squid-users] squid becomes very slow during peak hours

Thanks for replies,

1. i have tried squid 3.0 stable 14 for few weeks but the problems were there and performance issues was also severe. as we had previously 2.5 stable 10 running that's why i reverted to it temporarily. further i have squid 3.0/14 in place as i have install 2.5 in separate directry and i can squid 3.0/14 run it anytime. i will also welcome if you tell me the most stable version of squid.

2. secondly we are using RAID 5 and have very powerfull machine at present as compared to previous one, and previous was working good with the same amount of traffic and less powerfull system.

3. thirdly i have gigabit network card but yes i have 100 mb ethernet channel, but as defined in step 2 same link was working superb in previous setup.

4. i could not get chris robertson question regarding processors, i have two dual core xeon processors(3.2 ghz) and i captured stats at peak hours when performance was degraded.

So what should i do???

Regards,

--- On Wed, 7/1/09, Chris Robertson <crobertson_at_gci.net> wrote:

> From: Chris Robertson <crobertson_at_gci.net>
> Subject: Re: [squid-users] squid becomes very slow during peak hours
> To: squid-users_at_squid-cache.org
> Date: Wednesday, July 1, 2009, 2:25 AM
> goody goody wrote:
> > Hi there,
> >
> > I am running squid 2.5 on freebsd 7,
>
> As Adrian said, upgrade. 2.6 (and 2.7) support kqueue
> under FreeBSD.
>
> > and my squid box respond very slow during peak
> hours. my squid machine have twin dual core processors, 4
> ram and following hdds.
> >
> > Filesystem Size
> Used Avail Capacity Mounted on
> > /dev/da0s1a 9.7G 241M
> 8.7G 3% /
> > devfs 1.0K
> 1.0K
> 0B 100% /dev
> > /dev/da0s1f 73G
> 35G 32G
> 52% /cache1
> > /dev/da0s1g 73G
> 2.0G 65G
> 3% /cache2
> > /dev/da0s1e 39G
> 2.5G 33G
> 7% /usr
> > /dev/da0s1d 58G
> 6.4G 47G 12%
> /var
> >
> >
> > below are the status and settings i have done. i need
> further guidance to improve the box.
> >
> > last pid: 50046; load averages:
> 1.02, 1.07, 1.02
>
>
> up
> >
> > 7+20:35:29 15:21:42
> > 26 processes: 2 running, 24 sleeping
> > CPU states: 25.4% user, 0.0% nice, 1.3%
> system, 0.8% interrupt, 72.5% idle
> > Mem: 378M Active, 1327M Inact, 192M Wired, 98M Cache,
> 112M Buf, 3708K Free
> > Swap: 4096M Total, 20K Used, 4096M Free
> >
> > PID USERNAME THR
> PRI NICE SIZE RES STATE
> C TIME WCPU COMMAND
> > 49819 sbt 1 105
> 0 360M 351M
> CPU3 3 92:43 98.14% squid
> > 487 root
> 1 96 0 4372K
> 2052K select 0 57:00 3.47% natd
> > 646 root
> 1 96 0 16032K 12192K select
> 3 54:28 0.00% snmpd
> >
> SNIP
> > pxy# iostat
> > tty
> da0
> pass0
> cpu
> > tin tout KB/t tps
> MB/s KB/t tps MB/s us ni sy in
> id
> > 0 126
> 12.79 5
> 0.06 0.00 0
> 0.00 4 0 1 0 95
> >
> > pxy# vmstat
> > procs memory
> page
> disks
> faults cpu
> > r b w avm
> fre flt re pi po
> fr sr da0
> pa0 in sy cs
> us sy id
> > 1 3 0 458044 103268
> 12 0 0 0
>
> 30 5 0 0
> 273 1721 2553 4 1 95
> >
>
> Those statistics show wildly different utilization.
> The first (top, I
> assume) shows 75% idle (or a whole CPU in use). The
> next two show 95%
> idle (in effect, one CPU 20% used). How close (in
> time) were the
> statistics gathered?
>
> >
> > some lines from squid.conf
> > cache_mem 256 MB
> > cache_replacement_policy heap LFUDA
> > memory_replacement_policy heap GDSF
> >
> > cache_swap_low 80
> > cache_swap_high 90
> >
> > cache_dir diskd /cache2 60000 16 256 Q1=72 Q2=64
> > cache_dir diskd /cache1 60000 16 256 Q1=72 Q2=64
> >
> > cache_log /var/log/squid25/cache.log
> > cache_access_log /var/log/squid25/access.log
> > cache_store_log none
> >
> > half_closed_clients off
> > maximum_object_size 1024 KB
> >
> > if anyother info required, i shall provide.
> >
>
> The types (and number) of ACLs in use would be of interest
> as well.
>
> > Regards,
> > .Goody.
> >
>
> Chris
>
>

      
Received on Wed Jul 01 2009 - 18:41:21 MDT