Cloud router to onprem Cisco - BGP keeps going down

[shakalakka]

Hi,

Have a cloud router running BGP over VPN to onprem Cisco-routers. At random times i get a %BGP-3-NOTIFICATION: received from neighbor x.x.x.x 6/2 (Administrative Shutdown) 0 bytes. Then the session is brought back up and back to normal for some more days without interruption. Cant see any clues in the cloud router logs as to why this keeps happning. Anyone experienced something similar? Im not running BFD on these peerings, as im concerned it will get even worse.

[diannemcm · Accepted solution]

Hi @shakalakka,

Welcome to Google Cloud Community!

The following are potential root causes for BGP session flapping:

• MTU configuration on a BGP peering can disrupt the connection 

          Please see: On-premises router experiences BGP flap, MTU Considerations, Cloud Router BGP connection disconnects frequently

• Potentially triggered by the other side of the connection, which did not respond to any keepalives within the hold timeout (default 180 seconds). Check: BGP timers and settings.

Also, just to highlight this possible workaround: 

Graceful restart prevents BGP sessions from packet drops and route withdrawal during Cloud Router maintenance. If your on-premises device supports BGP graceful restart, enable it and set the graceful restart and stalepath timers as shown in the table:

If the issue still persists and you need further assistance, you can file a ticket with our support team.

I hope this helps!

[shakalakka]

Seems like youre spot-on with the maintenance windows at Google. Got it confirmed in a ticket. Thank you!