Being somewhat of a paranoid body, I have all my BGP peers currently set for
max-prefix of 75000. On Friday, my Sprint peer (usually around 67k)
exceeded that number and we stopped the session. This in itself is not a
big deal, but looking at http://www.telstra.net/ops/bgptable.html I notice
that they suck a large influx in the table at a similar time.. Did anyone
else see something similar? If so, what peers did you see this from? Here
is the log for my incident in question..
Dec 10 15:03:54 No. of prefix received from SPRINTROUTER reaches 67117, max 75000
Dec 10 15:31:16 No. of prefix received from SPRINTROUTER reaches 67137, max 75000
Dec 10 15:50:50 No. of prefix received from SPRINTROUTER reaches 67133, max 75000
Dec 10 15:54:10 No. of prefix received from SPRINTROUTER reaches 67100, max 75000
Dec 10 15:54:40 No. of prefix received from SPRINTROUTER reaches 67102, max 75000
Dec 10 15:55:10 No. of prefix received from SPRINTROUTER reaches 67099, max 75000
Dec 10 16:27:34 No. of prefix received from SPRINTROUTER reaches 67024, max 75000
Dec 10 16:42:38 No. of prefix received from SPRINTROUTER reaches 67024, max 75000
Dec 10 16:45:53 No. of prefix received from SPRINTROUTER reaches 67068, max 75000
Dec 10 17:18:58 No. of prefix received from SPRINTROUTER reaches 67078, max 75000
Dec 10 17:39:35 No. of prefix received from SPRINTROUTER reaches 73712, max 75000
Dec 10 17:39:39 No. of prefix received from SPRINTROUTER: 75001 exceed limit 75000
My session was disabled so I'm not sure how much bigger their table got.
The only reason I bring this up here instead of directly to Sprint is that
the data on the Telestra page appears to point the finger at this being a
multi-provider incident, and I haven't heard anything on this list or others
regarding the situation..
TIA,
-- Tim