Hi,
I am looking to setup a "backup" server in a separate DC from our main
cluster which should take the load of our site in the event of a
catastrophic failure of our active cluster.
This machine is therefore less powerful than the combined active
cluster and therefore not able to handle the full dynamic load which
is a coldfusion/php site with a MySQL backed. What I want to do is
configure squid as a reverse proxy, and *force* it to cache everything
that gets sent out. The backup machine will have 6GB RAM and 4 SCSI
drives in RAID 10 so I trust that squid can be configured to take full
advantage of this, but I just don't know how.
Mainly my question is there a way of telling squid to ignore any
cache-control headers and cache all content for 1 hour, with the most
common content cached in RAM and the rest cached on disk?
Secondly, does anyone have any configuration advice for a setup such as this?
With many thanks,
Alex Davies
Received on Sat Nov 05 2005 - 12:19:04 MST
This archive was generated by hypermail pre-2.1.9 : Thu Dec 01 2005 - 12:00:09 MST