snmp get for pool member count differs between ltm partners

We are receiving a Nagios alert for pool membership on one node only in a synchronised pair of ltms. It reports 3 active members whilst its partner (correctly) says 4.

 

Both ltms display display full pool membership (4) via the GUI interface.

 

These direct snmpget queries show the discrepancy:

 

/tmp/PETE snmpget -c XXXXXXX -v 2c ltm_host1 '.1.3.6.1.4.1.3375.2.2.5.1.2.1.23.30.47.112.114.111.100.117.99.116.105.111.110.47.115.101.99.112.103.119.46.108.118.46.99.111.109.45.112.111.111.108' F5-BIGIP-LOCAL-MIB::ltmPoolMemberCnt."/production/secpgw.lv.com-pool" = INTEGER: 3

 

/tmp/PETE snmpget -c XXXXXXX -v 2c ltm_host2'.1.3.6.1.4.1.3375.2.2.5.1.2.1.23.30.47.112.114.111.100.117.99.116.105.111.110.47.115.101.99.112.103.119.46.108.118.46.99.111.109.45.112.111.111.108' F5-BIGIP-LOCAL-MIB::ltmPoolMemberCnt."/production/secpgw.lv.com-pool" = INTEGER: 4

 

Any suggestions as to why this should occur and more importantly how we get the snmp query to accurately report the pool members?