[c-nsp] Etherchannel load-balacing change on live network

Chuck Church chuckchurch at gmail.com
Fri Jul 8 12:56:25 EDT 2016


Well, yeah.  I realize that is the correct fix.  I thought I read at some
point he was concerned about downtime involved with it.  That's where I
would look to see about potential bugs.  In case there is one that tends to
blackhole all traffic on that port channel, or QOS breaks, or whatever.  I
personally have changed the hash algorithm on a 6500 during moderate traffic
flow, was just a blip.  Hopefully that is the OP's outcome too.

 

Chuck

 

From: Mark Tinka [mailto:mark.tinka at seacom.mu] 
Sent: Friday, July 08, 2016 11:37 AM
To: Chuck Church <chuckchurch at gmail.com>; 'Satish Patel'
<satish.txt at gmail.com>
Cc: 'Cisco Network Service Providers' <cisco-nsp at puck.nether.net>
Subject: Re: [c-nsp] Etherchannel load-balacing change on live network

 

 

On 8/Jul/16 17:34, Chuck Church wrote:

If it's that much of a concern, I would look at the release notes for a
'current' IOS version for that platform, and check the caveats for bugs that
might have involved your version and etherchannel stuff.  Or a search on the
bug tool, though that involves a little luck on guessing the right keyword.


If the OP is getting biased hashing now, he really has no choice but to
enable IP entropy.

I'm sure taking a small hit is better than dropping packets due to poor
spraying of traffic.

Mark.



More information about the cisco-nsp mailing list