From: Scott M. Livingston (scottl@sprinthosting.net)
Date: Sun Mar 09 2003 - 02:06:32 GMT-3
Kurt,
Maybe you could load up some MRTG to aid you in seeing where your links
are getting pounded if in fact they are. Pruning is always a wise
practice in my opinion.
My infrastructure is a mix of 100Mb and GIG-E, but I make certain that
my core is all GIG-E and channeled - BW baby!!
Without seeing the topology it sounds as if you might have some
bottlenecks in that build?? Maybe you can channel up some of those 100M
links and or risers until you can find out what your traffic patterns
are and what is getting pounded; or not.
Speaking from experience make sure you have your roots and designated
ports set where they need to be set.... You mentioned MSFC's - make sure
your L2 paths line up w/ your L3 HSRP too if in fact you are leveraging
HSRP in your net.
You can go all GIG next week, but the most important thing to know is
that traffic pattern. If the logical L2 is jacked you might end up
spending a lot of money on BW when it's the logical configurations that
need the attention. Have fun!
MHO
-scott
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Kurt Kruegel
Sent: Saturday, March 08, 2003 10:09 PM
To: ccielab@groupstudy.com
Subject: any problems mixing 100mb and gb trunks ?
i'd like to ask a switching question.
we have a switch fabric of
2 6500's in the core
mffc's on each
and gb and 100mb line cards.
we have 20 2820's 100mb trunk
20 cat5000's 100mb trunk
and we are integrating
some 3548's
and 82 3550's
at gb to the core.
are there any potential issues with mixing 100/mb and 1000/mb trunks in
the
network without pruning ?
i have been noticing that cat5000's see a lot of in-discards
when i looked it up it seemed to say that they are frames that could not
be
switched.
also on some100/mb trunks off a 3548 i see giants and input errors.
the gb trunks show no problems.
could the 100mb be "choke points" until we go all gb ?
thanks.
This archive was generated by hypermail 2.1.4 : Sat Apr 05 2003 - 08:51:35 GMT-3