Hello,
I just finished the migration of our Zabbix server in version 2.2.1 from 2.0.8. Since this migration, my SNMP trafic from our Zabbix server is higher. I use a lot of SNMP items with dynamic indexes. Generally this is useful for the status of interfaces. Example :
Admin status port port1 the snmp_oid is .1.3.6.1.2.1.2.2.1.7[index,.1.3.6.1.2.1.2.2.1.2,port1]
I read this in documentation
, but when I sniff the trafic, I see that the browsing of the SNMP indexe tree are made on all check. It's like the cache doesn't work or the poller changed at any check.
I have set up 100 pollers in the configuration.
Maybe I make something wrong.
Other people have the same behaviour ?
Regards
Jean
I just finished the migration of our Zabbix server in version 2.2.1 from 2.0.8. Since this migration, my SNMP trafic from our Zabbix server is higher. I use a lot of SNMP items with dynamic indexes. Generally this is useful for the status of interfaces. Example :
Admin status port port1 the snmp_oid is .1.3.6.1.2.1.2.2.1.7[index,.1.3.6.1.2.1.2.2.1.2,port1]
I read this in documentation
"When a dynamic index item is requested, Zabbix retrieves and caches whole SNMP table under base OID for index, even if a match would be found sooner. This is done in case another item would refer to the same base OID later - Zabbix would look up index in the cache, instead of querying the monitored host again. Note that each poller process uses separate cache."
I have set up 100 pollers in the configuration.
Maybe I make something wrong.
Other people have the same behaviour ?
Regards
Jean
Comment