This article describes a condition where SRX High End is unable to generate ARP Request
Steps to Resolve this Problem: 1. "show route" shows that there's only ONE Route i.e. Directly connected:
{primary:node0}
root@srx3600-cluster> show route 10.20.0.31
inet.0: 11 destinations, 12 routes (11 active, 0 holddown, 0 hidden)
+ = Active Route, - = Last Active, * = Both
10.20.0.0/20 *[Direct/0] 05:28:52
> via reth0.1125
2. reth0.1125 is the interface that is directly connected to the network 10.20.0.0/20:
root@srx3600-cluster> show interfaces terse | match reth0.1125
xe-1/0/1.1125 up up aenet --> reth0.1125
xe-14/0/1.1125 up up aenet --> reth0.1125
reth0.1125 up up inet 10.20.0.1/20
3. FXP interface had the same ip address but it is currently
DOWN, so we assume it is NOT important:
{primary:node0}
root@srx3600-cluster> show interfaces terse | match fxp
fxp0 up up
fxp0.0 up down inet 10.20.0.1/20
4. But look here, in the Forwarding Table, we found two Routes to the same destination. One through FXP and the other through reth0.1125:
root@srx3600-cluster> show route forwarding-table destination 10.20.0.31
Routing table: default.inet
Internet:
Destination Type RtRef Next hop Type Index NhRef Netif
10.20.0.0/20 intf 0 rslv 332 1 fxp0.0
10.20.0.0/20 intf 0 rslv 517 1 reth0.1125
SRX prefers the route through the FXP interface even though it is down, and that's why we don't see the ARP generated from the reth0.1125 interface.
Note: FXP Interface is always going to be prefered over Data Plane Revenue port if a packet is originated from the SRX.