diff options
| author | Donatas Abraitis <donatas@opensourcerouting.org> | 2025-03-25 17:20:56 +0200 |
|---|---|---|
| committer | ton31337 <3352707+ton31337@users.noreply.github.com> | 2025-03-28 15:07:11 +0000 |
| commit | 405e413e6acb2f9239078a14e451a3ff529a90ef (patch) | |
| tree | 5c04eb5f427014ed4f3aa092ea4134f0a638a2f5 | |
| parent | 10d922b6316034a63c820e2c0e81c96dafaebff6 (diff) | |
bgpd: Retain the routes if we do a clear with N-bit set for Graceful-Restart
On receiving side we already did the job correctly, but the peer which initiates
the clear does not retain the other's routes. This commit fixes that.
Fixes: 20170775da3a3c5d41aba714d0c1d5a29b0da61c ("bgpd: Activate Graceful-Restart when receiving CEASE/HOLDTIME notifications")
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
(cherry picked from commit b7c657d4e065f310fcf6454abae1a963c208c3b8)
| -rw-r--r-- | bgpd/bgp_packet.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/bgpd/bgp_packet.c b/bgpd/bgp_packet.c index 9ebb2b21be..482e905636 100644 --- a/bgpd/bgp_packet.c +++ b/bgpd/bgp_packet.c @@ -1042,6 +1042,13 @@ static void bgp_notify_send_internal(struct peer_connection *connection, /* Add packet to peer's output queue */ stream_fifo_push(connection->obuf, s); + /* If Graceful-Restart N-bit (Notification) is exchanged, + * and it's not a Hard Reset, let's retain the routes. + */ + if (bgp_has_graceful_restart_notification(peer) && !hard_reset && + CHECK_FLAG(peer->sflags, PEER_STATUS_NSF_MODE)) + SET_FLAG(peer->sflags, PEER_STATUS_NSF_WAIT); + bgp_peer_gr_flags_update(peer); BGP_GR_ROUTER_DETECT_AND_SEND_CAPABILITY_TO_ZEBRA(peer->bgp, peer->bgp->peer); |