Tony Demark wrote:
> Is there a way to accelerate a set of websites with a
> Squid "appliance" machine sitting in front of N number of webservers with only
> one instance of squid on the "appliance"?
Of course there is. A single Squid can act as a accelerator for X
backendservers with Y IP-addresses and Z domains, and as a proxy and or
transparent proxy at the same time, without ambuiguity.
> There doesn't seem to be a way in 1.1 and I can't find any documentation to
> that effect on 2.1 or 2.2.
The only thing lacking from 1.1.X is proper Host: header support if you
need to support virtual Host: based servers. If you only have one domain
per IP then 1.1.X supports this fine.
The magic tool used in both versions are a redirector. When Squid
receives a virtually accelerated request without Host: header it inserts
the IP address of the interface receiving the request as host part in
the URL. This constructed URL is then sent to the redirectors for
rewriting.
> (2) Sibling relationship on accelerated servers
>
> Are there any problems with setting two squids that are in front of two
> identically configured (but different IP addressed) servers to treat each
> other as siblings?
No problems not seen when configuring two proxy-cache servers as
siblings. In fact the situation is almost identical.
Due to how Squid <2.2 works you should add some simple rules to prevent
cache loops. This is done by using cache_peer_access to deny requests
received from a sibling from being sent to a sibling. These loops are
quite harmless but clutters up the log files and may cause a tiny
performance degration.
---
Henrik Nordstrom
Spare time Squid hacker
