> I found that 3 or 4 PM-2E-30s will crater a 10BaseT collision segment.
Name Ipkts Ierrs Opkts Oerrs Collis Resets Queue
ether0 1303967 603 943559 4 165764 5 0
(165764/943559)*100
17.5679528200
Cool...17.6% colls according to the PM. I don't know if moving from RIP
to OSPF really did it any good, but last time I checked (when it was still
doing RIP) it was just over 20%.
This is a segment with a PM2ei20, 5.3 PM2e30's worth of linux term servers
(3 big linux term servers), a news server, tape server, and dozen or more
other boxes.
> Before spending the money for a router I'd first try getting a switching
> hub for that installation.
Our switches should arrive later this week :) Due to the size of the
switches I ordered, I'm going to have to have 2 "segments" with multiple
systems...one will likely have several term servers, one will have generic
boxes. Our internet gateway, news server, mail server, IRC server, tape
server, and big virtual web server will all get their own switch ports.
Hopefully, just separating the news server from the PM will seriously
reduce the PM's "claimed" collision rate.
------------------------------------------------------------------
Jon Lewis <jlewis@fdt.net> | Unsolicited commercial e-mail will
Network Administrator | be proof-read for $199/hr.
________Finger jlewis@inorganic5.fdt.net for PGP public key_______