Re: [squid-users] High CPU utilization

From: Adrian Chadd <adrian_at_squid-cache.org>
Date: Tue, 28 Jul 2009 09:50:46 +0800

Change "ufs" to "aufs" - assuming you compiled in aufs.

Consider upgrading to Squid-2.7.STABLEx - I did a whole lot of little
performance tweaks between 2.6 and 2.7.

Learn about oprofile and submit some performance information to help
developers. :)

Adrian

2009/7/28 jotacekm <minuzzo_at_viaip.com.br>:
>
> Hello.
> Recently we have added a lot more clientes behind a squid proxy, and now cpu
> utilizitaion is usually 70-95%. The processor is a intel dual core 2160  @
> 1.80GHz. Users started complaing about the speed on accessing pages, and the
> link is fine.
>
> Here is squidclient mgr:info:
>
> Squid Object Cache: Version 2.6.STABLE5
> Start Time:     Fri, 24 Jul 2009 20:19:16 GMT
> Current Time:   Mon, 27 Jul 2009 19:09:07 GMT
> Connection information for squid:
>        Number of clients accessing cache:      1561
>        Number of HTTP requests received:       8590404
>        Number of ICP messages received:        0
>        Number of ICP messages sent:    0
>        Number of queued ICP replies:   0
>        Number of HTCP messages received:       0
>        Number of HTCP messages sent:   0
>        Request failure ratio:   0.00
>        Average HTTP requests per minute since start:   2021.3
>        Average ICP messages per minute since start:    0.0
>        Select loop called: 122560206 times, 2.081 ms avg
> Cache information for squid:
>        Request Hit Ratios:     5min: 23.0%, 60min: 21.9%
>        Byte Hit Ratios:        5min: 13.3%, 60min: 14.7%
>        Request Memory Hit Ratios:      5min: 16.3%, 60min: 17.0%
>        Request Disk Hit Ratios:        5min: 26.3%, 60min: 29.9%
>        Storage Swap size:      4609784 KB
>        Storage Mem size:       65692 KB
>        Mean Object Size:       15.67 KB
>        Requests given to unlinkd:      455490
> Median Service Times (seconds)  5 min    60 min:
>        HTTP Requests (All):   1.24267  1.46131
>        Cache Misses:          1.54242  1.81376
>        Cache Hits:            0.28853  0.30459
>        Near Hits:             1.31166  1.62803
>        Not-Modified Replies:  0.23230  0.23230
>        DNS Lookups:           0.29097  0.31806
>        ICP Queries:           0.00000  0.00000
> Resource usage for squid:
>        UP Time:        254991.358 seconds
>        CPU Time:       56029.670 seconds
>        CPU Usage:      21.97%
>        CPU Usage, 5 minute avg:        94.38%
>        CPU Usage, 60 minute avg:       93.36%
>        Process Data Segment Size via sbrk(): 193008 KB
>        Maximum Resident Size: 0 KB
>        Page faults with physical i/o: 70
> Memory usage for squid via mallinfo():
>        Total space in arena:  193008 KB
>        Ordinary blocks:       179871 KB   6058 blks
>        Small blocks:               0 KB      0 blks
>        Holding blocks:          1080 KB      2 blks
>        Free Small blocks:          0 KB
>        Free Ordinary blocks:   13136 KB
>        Total in use:          180951 KB 93%
>        Total free:             13136 KB 7%
>        Total size:            194088 KB
> Memory accounted for:
>        Total accounted:       111416 KB
>        memPoolAlloc calls: 973288506
>        memPoolFree calls: 972077571
> File descriptor usage for squid:
>        Maximum number of file descriptors:   4096
>        Largest file desc currently in use:   1604
>        Number of file desc currently in use: 1308
>        Files queued for open:                   0
>        Available number of file descriptors: 2788
>        Reserved number of file descriptors:   100
>        Store Disk files open:                  30
>        IO loop method:                     epoll
> Internal Data Structures:
>        300182 StoreEntries
>         14733 StoreEntries with MemObjects
>         14414 Hot Object Cache Items
>        294106 on-disk objects
>
> And here is part of of squid.conf:
>
> http_port 3128
> visible_hostname xxx
> hierarchy_stoplist cgi-bin ?
> acl QUERY urlpath_regex cgi-bin \?
> cache deny QUERY
> acl apache rep_header Server ^Apache
> broken_vary_encoding allow apache
> access_log /var/log/squid/access.log squid
> cache_store_log none
> hosts_file /etc/hosts
>
>
> #
> ----------------------------------------------------------------------------------------------------------
> cache_mem 64 MB
> cache_dir ufs /var/spool/squid 5000 50 256
> cache_replacement_policy heap LFUDA
> maximum_object_size 51200 KB
> maximum_object_size_in_memory 64 KB
> memory_replacement_policy heap GDSF
> logfile_rotate 3
>
>
> #
> ----------------------------------------------------------------------------------------------------------
> refresh_pattern ^ftp:           1440    20%     10080
> refresh_pattern ^gopher:        1440    0%      1440
> refresh_pattern .               0       20%     4320
> acl all src 0.0.0.0/0.0.0.0
> acl manager proto cache_object
> acl localhost src 127.0.0.1/255.255.255.255
> acl to_localhost dst 127.0.0.0/8
>
>
>
> max_open_disk_fds 2046
>
> # timeouts
> connect_timeout 30 seconds
> shutdown_lifetime 5 seconds
> forward_timeout 2 minutes
> pconn_timeout 30 seconds
> persistent_request_timeout 1 minute
> request_timeout 2 minute
>
> Is there anything that i can do to lower the cpu utilization, or do i  have
> to upgrade the hardware?
> Thanks.
>
>
> --
> View this message in context: http://www.nabble.com/High-CPU-utilization-tp24686629p24686629.html
> Sent from the Squid - Users mailing list archive at Nabble.com.
>
>
Received on Tue Jul 28 2009 - 01:50:56 MDT

This archive was generated by hypermail 2.2.0 : Tue Jul 28 2009 - 12:00:05 MDT