Re: [squid-users] cache_peer configuration is not being reliably hit in Squid 3.2.1

From: Amos Jeffries <squid3_at_treenet.co.nz>
Date: Thu, 20 Sep 2012 16:30:48 +1200

On 20/09/2012 3:53 p.m., Nathan Hoad wrote:
> Hello all,
>
> I'm running Squid 3.2.1, with a cache_peer configured to point at
> another local Squid instance as a parent proxy. The parent proxy is
> not getting reliably hit, for the *exact* same url.
>
> acl objectcache url_regex -i "/path/to/regexes"
>
> cache_peer localhost parent 60084 0 proxy-only
> name=objectcachepeer-peer no-digest no-query
> cache_peer_access objectcache-peer allow objectcache
> client_dst_passthru off

 From your use of client_dst_passthru I assume you have this Squid
configured as a transparent interception proxy?

> Example output from access.log:
>
> 10.3.100.40 "-" [20/Sep/2012:10:46:21 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 8898496 TCP_MISS_ABORTED:HIER_DIRECT - k1
> 10.3.100.40 "-" [20/Sep/2012:10:46:34 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 2903696 TCP_MISS_ABORTED:HIER_DIRECT - k1
> 10.3.100.40 "-" [20/Sep/2012:10:46:37 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 2290496 TCP_MISS_ABORTED:HIER_DIRECT - k2
> 10.3.10.178 "-" [20/Sep/2012:10:48:03 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 13221762 TCP_MISS_ABORTED:FIRSTUP_PARENT - k1
> 10.3.10.178 "-" [20/Sep/2012:10:48:22 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 76987317 TCP_MISS_ABORTED:FIRSTUP_PARENT - k1
> 10.3.10.178 "-" [20/Sep/2012:10:48:26 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 4117496 TCP_MISS_ABORTED:HIER_DIRECT - k1
> 10.3.10.178 "-" [20/Sep/2012:10:48:30 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 4228096 TCP_MISS_ABORTED:HIER_DIRECT - k1
> 10.3.100.40 "-" [20/Sep/2012:10:48:56 +1000] "GET
> http://appldnld.apple.com/iOS6/Restore/041-7181.20120919.lEuOK/iPhone4,1_6.0_10A403_Restore.ipsw
> HTTP/1.1" 200 11354096 TCP_MISS_ABORTED:HIER_DIRECT - k1

This is apache format for web end-servers. Do you have anything in Squid
format which has better logging information for proxy details?

Any cache.log traces of LIVE/DEAD status on the peer and indication of
how that interacts with the access.log timing? It could simply be the
peer going unavailable for a while.

>
> This issue was also present in 3.2.0.19 and 3.2.0.20. Are there any
> configuration options I may be missing? I've had this working before,
> in Squid 3.2.0.16 and Squid 3.2.0.17, with the options detailed above.
>
> Thanks,
>
> Nathan.
Received on Thu Sep 20 2012 - 04:31:01 MDT

This archive was generated by hypermail 2.2.0 : Thu Sep 20 2012 - 12:00:04 MDT