SNMP Timemout

[leoh]

I have 160 devices monitored in zabbix with SNMP

Everyday I got more then 10 snmp agents offline (device:161 connection timeout)

SNMP is working (snmpwalk ok), community is working since it works for a while and then stops.

Questions:
1 - How to set SNMP Timeout so that zabbix does not give up so easily?
2 - How do I re-enable SNMP on those devices? Zabbix seems to give up forever. Devices get unmonitored and they are never restored to its previous snmp-online state!

[leoh]

I have 160 devices monitored in zabbix with SNMP

Everyday I got more then 10 snmp agents offline (device:161 connection timeout)

SNMP is working (snmpwalk ok), community is working since it works for a while and then stops.

Questions:
1 - How to set SNMP Timeout so that zabbix does not give up so easily?
2 - How do I re-enable SNMP on those devices? Zabbix seems to give up forever. Devices get unmonitored and they are never restored to its previous snmp-online state!

I saw that it will be fixed in Zabbix 2.2

https://support.zabbix.com/browse/ZBX-4393

The problem is the LLD. I think that zabbix tries to snmpwalk the whole MIB and, in low connections, it exceeds the timeout. In zabbix 2.2 SNMP timeout will be set in zabbix_server.cfg (and net-snmp will respect this setting)

By know I will just turn off Low Level Discovery of my snmp devices so that zabbix does not disable monitoring just because of this tiny bug.