]> git.puffer.fish Git - matthieu/frr.git/commit
bgpd: Upon interface up (update) only kick-off non-Established peers
authorvivek <vivek@cumulusnetworks.com>
Thu, 18 Aug 2016 22:02:49 +0000 (15:02 -0700)
committervivek <vivek@cumulusnetworks.com>
Thu, 18 Aug 2016 23:50:09 +0000 (16:50 -0700)
commite5d1e72daaf5775e4bf67d812553d4fd8775791a
tree699f48480bf052457a53005182055ec0ffc7ea3b
parent926ea62e0bfb598bfa2fb950e28861b2588f41ea
bgpd: Upon interface up (update) only kick-off non-Established peers

Any interface flags/parameter change (e.g., MTU, PROMISC flag change) is
notified by zebra to clients as an "up" event. BGP literally treats this
as the interface coming up and kicks all neighbors on that interface (i.e.,
directly connected peers). When doing so for IPv4 peers on the interface
(numbered or unnumbered /30-/31) or IPv6 numbered peers, peers that may
already be Established are also flapped; when doing so for IPv6 unnumbered
peers (classic 'neighbor swpX interface' scenario with no configured IP
address on interface), only peers not in Established state are processed.

This patch fixes the code to ensure that in all cases, only non-Established
peers are kicked.

Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
Reviewed-by: Don Slice <dslice@cumulusnetworks.com>
Reviewed-by: Chris Cormier <chriscormier@cumulusnetworks.com>
Ticket: CM-12526
Reviewed By: CCR-5119
Testing Done: Manual, bgp-min
bgpd/bgp_nexthop.c