Re: multiple squids re-visited

From: Clifton Royston <cliftonr@dont-contact.us>
Date: Mon, 10 Jan 2000 11:53:19 -1000

On Mon, Jan 10, 2000 at 12:23:48PM -0800, Jimi wrote:
> I've been changing around the configuration of a bank of squids, all of
> which have fairly small caches. I have them in digest mode, all connected
> to a parent squid, which is configured with the proxy-only option. The
> siblings don't actually know about each other...only the parent. I do this
> for configuration reasons, since the squids being on or off the network is
> fairly dynamic, and having to re-configure the 'squid.conf' files would be
> tedious.
>
> The way I expect this to work, is that the parent will receive digests from
> all the siblings, and then if a sibling requests an item that another
> sibling has, it can get it through the parent. Should this work? It
> doesn't seem to. I see the sibling request pages that I know another
> sibling has, and it goes through the parent who goes directly to get the
> page. Maybe they aren't sharing digest information? Maybe the parent will
> never ask a sibling for information? This is my guess, but I thought I
> would ask some more knowledgeable people.

The definition of a parent is that requests will only flow "upward"
from a child to parent, never from the parent to the child, but
children are allowed to ask the parent for objects they do not have in
cache. With a sibling relationship between two caches, the sibling is
normally allowed to get objects that the other has in its cache, but
not to request objects which it does not have. (In other words, your
siblings are not really siblings, since they don't talk directly to
each other, just share a common parent. Half-brothers? ;-) )

It *might* be possible to make it work the way you want, if you
configure the parent cache so that it sees all of the other "children"
as its siblings (in which case it will ask them for things which it
knows they have) but configure all of the children to treat it as a
parent. It may take some playing around with, as it doesn't sound like
an expected configuration.

  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr@lava.net
        "An absolute monarch would be absolutely wise and good.  
           But no man is strong enough to have no interest.  
             Therefore the best king would be Pure Chance.  
              It is Pure Chance that rules the Universe; 
          therefore, and only therefore, life is good." - AC
Received on Mon Jan 10 2000 - 15:04:04 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:50:18 MST