Description

This articles explains the scenario when secondary-ip-address is NOT configured in IP-Monitoring, IP monitoring packet is not sent out on RG+ secondary node.

Symptoms

According to Example: Configuring IP Monitoring on SRX Series Devices , it is typical to configure a secondary IP address in IP-monitoring such that the secondary IP address is used to send ICMP packets from the secondary node to check the reachability of the monitored IP.

Below is a configuration example with interface and secondary-ip-address configured:

{primary:node0}[edit]
user@host# set chassis cluster redundancy-group 1 ip-monitoring family inet 25.25.25.254 weight 80
user@host# set chassis cluster redundancy-group 1 ip-monitoring family inet 25.25.25.254  interface reth0.0 secondary-ip-address 25.25.25.1

In the case when IP monitoring  is configured without secondary-ip-address and interface:

redundancy-group 1 {
    node 0 priority 200;
    node 1 priority 100;
    ip-monitoring {
        family {
            inet {
                25.25.25.254 weight 255;  <-- No secondary-ip-address configured
            }
        }
    }
}

A warning is shown saying, 'limited monitoring functionality' while committing configuration changed.

[edit chassis cluster redundancy-group 1 ip-monitoring family inet]
  '25.25.25.254'
    Warning: interface option is not configured. You might get limited monitoring functionality

In the case when monitoring IP "25.25.25.254 " is not reachable from secondary node(Node1) of RG1,  IP Monitoring does not trigger a RG1 failover  and priority of Node1 remains 100.  

root@srx# run show chassis cluster status    
...
Redundancy group: 1 , Failover count: 1
    node0                   200         primary        no       no  
    node1                   100         secondary      no       no 

Solution

This is because IP monitoring does not work on the secondary node as no interface and secondary-ip-address is configured.  This is the "limited monitoring functionality" while committing configuration changed. Hence, to enable IP-monitoring in the secondary node, it is necessary to configure secondary-ip-address. 

When no secondary-ip-address is configured,  it is not possible to tell whatever the target IP is reachable or not.  In the example below, we have vty CLI from RG1+ secondary node. Even if monitored IP is reachable and Rx count is increasing, IP monitoring packet is not sent out.

[flowd]FPC6.PIC0(vty)# sh usp ha ip-mon status    
Monitored IPs
----------------------------------------------------------
Index: 0
        IP: 25.25.25.254 (len: 4), ver: 1, Status: reachable ---> Status remains reachable
        ICMP Tx count: 56842, Rx count: 56842 , Seq Num: 336 ---> Rx continues increasing
        Gateway IP: 25.25.25.254 (len: 4)
        Secondary IP: 25.25.25.2 (len: 4), Changed: Yes
        Interval: 1, threshold: 5
        Routing-table: 0
        Secondary ifl idx: 0
        Secondary mac addr: 00:00:00:00:00:00/48
        Reth info count: 0
        Reths:
        RG State: Master
        Failure reason: 0