Andrew’s theory about uplinks is correct:
Ridgeview 3/35 and Po1 are part of its link to 511-7804
And 511-7504 po11 and 6/36 are part of its link to 511-7804
I don’t have any theories but spent a little time looking around, my notes so far:
The usual background of non-unicast packets jumped from ~100/sec to ~250/sec rom 14:10 to 15:01
They were egressing lots of ports but po4 (Charter) is the only ingress port I’ve found so far - but I don’t think it’s volume accounts for it all.
A number of members had traffic volumes change but the only ones who stopped all together that I see are:
Integris and Hoyos on your remote
WiscNet on their remote
Conversely Xcel Entergy had a spike in traffic at around that time.
And then we had trouble graphing NOCIX remote ports during that time too, which is odd.
From:
Andrew Hoyos <hoyosa@gmail.com>
Date: Wednesday, April 30, 2025 at 4:08 PM
To: Jay Hanke <jayhanke@southfront.io>
Cc: Sergii Polishchuk <sergi@infotec.ca>, Ryan Nelson <ryan.nelson@integrisit.com>, "mice-discuss@lists.micemn.net" <mice-discuss@lists.micemn.net>
Subject: Re: MICE connectivity issues / elevated multicast traffic
We are, but saw issues to all peers (non BFD peers as well)
There was definitely some uptick in non-unicast traffic in a few spots. I’m sure some are switch to switch links...
7804 - 3/2/1 -> ridgeview
http://micelg.usinternet.com/cacti/graph.php?action=view&local_graph_id=4027&rra_id=all
-> Ridgeview 3/35/1 + po1
7504 - po11 + 6/36/1
http://micelg.usinternet.com/cacti/graph.php?action=view&local_graph_id=1859&rra_id=all
Andrew Hoyos
hoyosa@gmail.com