Re: What's the best configuration for this setup?

From: Ross Wheeler <rossw@dont-contact.us>
Date: Sun, 24 Nov 1996 17:54:58 +1100 (EST)

> Whoever started this thread (sorry I forgot!) claimed that a single
> (FTP?) transfer seemed to consume the entire link bandwidth. TCP is
> supposed to have this nice load-sharing algorithm, so its not clear to
> me how this could happen.

OK. No-one dobbed me in, so I better own up :-)

Just for the record, it came about after I'd been running squid on a
local server for all the locally attached (ethernet) machines (I run an
internet Walk-in centre, amongst other things). The benefit of the cache
was immediately apparent. After running if for a while, I told our squid
about the squid upstream. Wham. Within 30 minutes, I figured something
was horribly wrong when I noticed some of the other things I was doing
_outside_ our domain became very slow. On tracking down usage, link
reponse times and traffic levels, that's when I identified it was squid
to squid traffic that was eating up all the bandwidth. It may NOT have
been a single transfer that did it, but the combined affect of multiple
requests. That, I didn't chase up. But: our squid is well connected (on
the same segment as our router), and the squid at the other end of the
ISDN 128K is also "well connected". They were transfering a bucket load
of data in short order. By the time link performance returned and I got
off the phone, I noticed "inter-squid" traffic was near nothing again.

Regards, RossW
Received on Sat Nov 23 1996 - 22:55:07 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:33:37 MST