On Wed, Apr 03, 2013 at 06:45:15AM -0500, David Farmer wrote:
> This morning at approximately 1:30 our peering session with HE
> reset, I assume for some kind of normal maintenance on their end.
>
> This triggered a weird issue on our end.  Because of some crufty
> route policy we were preferring the routes we learned from the MICE
> route server over the routes we learned directly from HE.  And for
> some reason the routes we were learning from the route server had a
> different link-local address associated with them than the routes we
> learned directly.  And this address no longer seems to be valid, and
> this caused traffic to be black holed.
>
> From below
> FE80::224:38FF:FEA7:EA03 - address on routes learned from MICE RS
> FE80::224:38FF:FEA7:EA36 - address on routes direct from HE

FIWW: We do buy transit from HE and did get the maintenance notice for
this morning.

I've also seen this behavior from them before. I'm not sure if the link
local address changes on their end due to some change or what is going
on. It doesn't matter on our direct connect router, where we only
talk the public IPv6 address, but the MICE connection seems to have
stale link-local IPv6 addresses from time to time that I've had to
clear out due to some change.

--
Doug McIntyre                            <[log in to unmask]>
          -- ipHouse/Goldengate/Bitstream/ProNS --
       Network Engineer/Provisioning/Jack of all Trades

########################################################################

To unsubscribe from the MICE-DISCUSS list, click the following link:
http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1