Guido Serassio wrote:
Hi,
From the provided release notes about Windows limitations:
* DISKD: still needs to be ported
* WCCP: cannot work because user space GRE support on Windows is missing
* Transparent Proxy: missing Windows non commercial interception driver
* Some code sections can make blocking calls.
* Some external helpers may not work.
* File Descriptors number hard-limited to 2048.
So, you cannot do transparent proxy on Windows.
Regards
Guido Serassio
Acme Consulting S.r.l.
Microsoft Gold Certified Partner
VMware Professional Partner
Via Lucia Savarino, 1 10098 - Rivoli (TO) - ITALY
Tel. : +39.011.9530135 Fax. : +39.011.9781115
Email: guido.serassio@xxxxxxxxxxxxxxxxx
WWW: http://www.acmeconsulting.it
-----Messaggio originale-----
Da: 4N0 [mailto:sinkohyo@xxxxxxxxx]
Inviato: martedì 13 aprile 2010 9.17
A: squid-users@xxxxxxxxxxxxxxx
Oggetto: Squid 2.7 port on Windows scenario
Hello,
I have a following scenario:
Squid on one frontend Windows server that is needed to serve as image
cache
from two backend windows servers running asp.net applications. Sqiud also
is
needed for intelligent switch in case any of backened servers is dead (if
server 2 is dead switch to server 3 and vice versa).
Worth noting that you do not need Squid to be running on windows.
For best performance one of the Linux or Unix OS is recommended over
Windows.
My question is, how I can achieve this scenario with squid on
configuration
and hardware level? I've read squid documentation, example scenarios but
can't get it to run. AFAIK my proxy needs to fulfill transparent proxy to
remote box scenario. But maybe also reverse proxy? (I only want to cache
static content, and "balance" switching).
You only need the reverse-proxy feature for the above requirements.
With a properly setup reverse-proxy everything goes through the Squid
gateway and what can be cached is (usually the static bits, but also
some dynamic as well sometimes). So transparent never gets in the way to
complicate things. This also ads the benefit of being able to scale out
easily by plugging more Squid in at the front end.
The basic config example in the wiki is what you want to start with.
http://wiki.squid-cache.org/ConfigExamples/Reverse/BasicAccelerator
For the failover requirement simply configure two cache_peer parent
entries. One for each of the back-end servers. With identical
cache_peer_access or cache_peer_domain settings.
For balancing there are a few methods, such as round-robin,
sourcehash, weighted or CARP to choose from. Pick the one that meets
your balancing needs and tune to suite your likings.
http://www.squid-cache.org/Doc/config/cache_peer
Amos
--
Please be using
Current Stable Squid 2.7.STABLE9 or 3.1.1
