Date: Sat, 03 Apr 1999 17:36:53 -0800 From: Dann Lunsford <dann@greycat.com> To: questions@freebsd.org Subject: Problem with L4 Switch Message-ID: <3706C235.66368C36@greycat.com>
next in thread | raw e-mail | index | archive | help
My ISP recently put up a squid (2.1, according to the messages I get
now), and they're using a
Foundry level-4 switch to divert all HTTP traffic to it. Problem is,
apparently FreeBSD (and Linux) tickle a bug in the switch: I see the
three-way
handshake between me and the remote system, then the initial GET and its
ACK, then absolutely *nothing* gets back to me until I terminate (and
then all I see is the FIN and its ACK). The only site this doesn't
happen with is, by some strange coincidence, the ISP's own website,
which is on *this* side of the switch.
Now I've got a strange mix of machines here (a FreeBSD box, a 3-node
VAXcluster, several Linux boxes, a couple of OS/2 nodes, and a couple of
PDP-11's :-)), and I *think* I see a pattern: The machines running a
BSD derived stack (the Linux boxes, the FreeBSD box, and one of the OS/2
systems) are all having the problem (HTTP packets
tossed by the switch), the other systems are not.
I was told by an acquaintance who works for the ISP that apparently a
few NT systems have the problem, also. SO...questions: Anyone else out
there seen something like this, is there a fix (other than getting the
ISP to throw out the switch, which has about as much chance as the
proverbial nitrocellulose dog in hell), or at least a workaround?
Thanks in advance.
Dann Lunsford
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3706C235.66368C36>