From: Notari, Ed (NotariE@usa.redcross.org)
Date: Thu Dec 12 2002 - 10:37:02 EST
Original question below.
I got one response and it was just what I was looking for. Rather than
paraphrase here is the response from Jesper Frank Nemholt...
===========================================================
Yes it's correct, and yes the order is (often) important.... at least with
NetRAIN configurations my experience is that using lan_config afterwards
doesn't work, not on the physical interfaces, nor on the virtual NetRAIN
interface....so it's likely the same with LAG.
An additional hurdle is that if you setup NetRAIN in rc.config, then it'll
be called during init from /sbin/rc3.d/S00fnetrain which is before
inet.local which is called from /sbin/rc3.d/S00inet, so when configuring
NetRAIN in rc.config, your lan_config statements in /etc/inet.local will be
executed too late and thus not work. One fix is to put all netRAIN into
/etc/inet.local and forget about the normal rc.config style setup. Another
is to create a 2nd init script that executes lan_config before the NeRAIN
init script ...but these problems are only related to NetRAIN, not LAG
(AFAIR).
--
Un saludo / Venlig hilsen / Regards
Jesper Frank Nemholt
Hewlett-Packard Company
-----Original Message-----
From: Notari, Ed [mailto:NotariE@usa.redcross.org]
Sent: Tuesday, December 10, 2002 4:02 PM
To: tru64-unix-managers@ornl.gov
Subject: lan_config and lag_config
I am planning to setup a link aggregation group (lag) on my ES40 and I was
wondering about how lan_config fits into the equation.
Previously we needed to add the following to inet.local...
/usr/sbin/lan_config -i ee0 -a 0 -s 100 -x 1 ... In order to get FULL DUPLEX
working.
When I add in two more ee# cards it would seem that I would want to have
similar lan_config statements... /usr/sbin/lan_config -i ee1 -a 0 -s 100 -x
1 /usr/sbin/lan_config -i ee2 -a 0 -s 100 -x 1
After which would come the lagconfig statements to join ee1 and ee2 into a
"lag0".
Is this correct and is the order of the statements in inet.local important
in this case?
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:01 EDT