Yep, CEF was the cause of this.
Configuring 'ip load-sharing per-packet' on the relevant interfaces has
evened up the load again. Now, of course, there is the risk that both
links will be flooded instead of just one, but never mind...
Thanks for all the suggestions.
JMcL
----- Forwarded by Jenny Mcleod/NSO/CSDA on 29/10/2001 11:02 am -----
"jenny.mcleod@centreli
nk.gov.au" To:
[EMAIL PROTECTED]
<jenny.mcleod
cc:
Sent by: Subject: fast
switching [7:23969]
[EMAIL PROTECTED]
24/10/2001 11:07
am
Please respond
to
"jenny.mcleod@centreli
nk.gov.au"
Something odd is happening on my network... (not that that's unusual...)
I have a couple of 7507s connected by two E1 links. For various reasons,
the links are set up with fast switching disabled (mainly because there are
single sessions with enough traffic to flood a single link). The 7507 on
one side is running IOS 11.2. Last weekend the 7507 on the other side was
upgraded to 12.1(10).
According to 'show ip int', fast switching and flow switching are disabled
(on both links at both ends). However, MRTG shows that the traffic from
the 12.1 router to the 11.2 router is not balanced evenly across the two
links. Traffic from the 11.2 router to the 12.1 router is balanced.
Any guesses as to why this is so? Bug (surely not, this is Cisco...:-)?
Is a reboot or shut/no shut required to change switching states (I didn't
do the upgrade myself and I'm not sure what exact configuration sequence
was used)? Something really obvious I'm missing here?
Ta,
JMcL
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=24442&t=23969
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
