future of icap-patch

Hi all.

My company uses squid with ntlm and icap. That's a very unlucky combination
I understand, as the history tells us that it's an unstable one.

I've managed to fix or work around the various assertions in 2.5.STABLE6, so
that we can be sure it's not crashing. You see, our farms are the frontend
for roughly 40K users these days and without my work it wasn't that unusual
for up to 70% of the servers to go down at once (round robin load balancer
outside our jurisdiction). I've taken a look at 2.5.STABLE9 and managed
sofar to apply all the patches but the 2Gig one to this combination,
creating a thing that crashes sometimes 3 times within one hour. For us
that's of course not acceptable. I do know roughly where the reason lies,
but I am not willing to invest any work into it as long as the patch is not
going to be part of the main squid code. Why? Because I have to go through
the same nightmare everytime a needed patch comes along.

I do understand that there were a couple of people in the past asking for
exactly that and it was denied. I am currently in the position to have the
freedom to start a new icap code project. But before I dive headfirst into
it, I'd ask if it is at all possible to make it a part of the mainline once
it reaches a certain maturity and if so, what do I have to do for it. If
there's no way anyway, I'd write it in a way convenient for me and my
company alone, skipping tedious tasks like keeping the code portable. In
such a case... What about building an interface into squid that allows to
add functionality like the ones icap offers? It'll prevent at least that any
additional code will have to break the rest of the code.

Regards, Oliver Baumgaertel
Received on Mon May 02 2005 - 01:58:31 MDT