On VMWare deployments, interface route may remain in down state due to a race condition during SE bring-up leading to traffic disruption.
C00 06:37:00.530201 [arpresolve:536] arpresolve: can't allocate llinfo for [10.x.x.x]
C00 06:37:00.530201 [ngx_event_connect_peer:194] *41 error [22] - connect() to 10.x.x.x failed
C00 06:37:00.530201 [ngx_http_upstream_finalize_request:5777] *41 Upstream Response: [10.x.x.x]:443 => [11x.11x.11x.xxx]:28527 => [10.x.x.x]:443 => [10.x.x.x]:51490 0 sent:0 recvd:0 r:0x5646cxxxxx c:0x5646xxxxx, upstream c:0x5646xxxxx
VMware
show serviceengine se-xbtcm route
+-----------------+----------------+-----------+---------------+---------------------+
| IP Destination | Gateway | Interface | Interface IP | Route Flags |
+-----------------+----------------+-----------+---------------+---------------------+
VRF 1
+-----------------+----------------+-----------+---------------+---------------------+
| 0.0.0.0/0 | 10.x.x.x | eth9 | 10.x.x.x | Up, Static,Gateway |
| 10.x.x.x/24 | 0.0.0.0 | eth9 | 10.x.x.x | Down, GWUp |
| 10.x.x.x/32 | 0.0.0.0 | eth9 | 10.x.x.x | Up, GWUp |
+-----------------+----------------+-----------+---------------+---------------------+
Temporary Workaround:
we can flap the interface by running the following commands from the namespace:
ifconfig avi_<eth_name> down
ifconfig avi_<eth_name> up
Permanent Fix:
Perform the upgrade to the following versions where the fix has been applied:
31.1.1
22.1.7
30.2.2
22.1.6-2p4