Re: [squid-users] WCCP becomes unusable

On Wed, 13 Feb 2002, Adrian Chadd wrote:

> On Wed, Feb 13, 2002, Gergely Madarasz wrote:
> > notified about the change or proxy1 doesn't correctly parse this
> > notification, so no reconfiguration happens). I really needed a fix for
> > this, or at least a workaround, so I modified the wccp code to reconfigure
> > the hash each 15 minutes (more exactly each 180 iseeyou messages :)), no
> > matter what happens.
>
> Right. Well, it seems that as it "loses" the cache (eg a log rotation
> which lasts a damned long time) the cache doesn't pick up its out of
> the hash assignment, and doesn't go and resubmit itself? Or it thinks
> its the designated cache (which assigns the caches into the hash buckets..)

I don't know exactly how wccp works, but from what I read from the squid
wccp code it seemed to me, that squid doesn't check anything in the
iseeyou messages except for the change member, so how could it see that
its out of the cache assignment?

And how should the cache resubmit itself? I only see that it sends hereiam
packets like it should, and the cisco router sees them, and show ip wccp
web-cache detail shows them as usable, there's just no hash assigned to
them (and when the designated cache reconfigures the hash without any
change notification, the cache gets his share)

Btw when I searched the web about this problem, I came to a page where
they wrote about a similar problem with the cisco cache engine. The
solution (hah!) explained there was to disable wccp for 30 seconds, then
enable it again on the cache engine. So it might be a problem with the
wccp protocol after all...

-- 
Madarasz Gergely          gorgo@sztaki.hu          gorgo@linux.rulez.org
    It's practically impossible to look at a penguin and feel angry.
        Egy pingvinre gyakorlatilag lehetetlen haragosan nezni.
                  HuLUG: http://mlf.linux.rulez.org/