Hi,
I am using checkmk enterprise edition. Monitoring network, linux and urls.
I have issues in monitoring network devices. Sometime it takes long to get the snmp data.
I noticed it is polling a bunch of default mibs for all devices. Is there any way I can customize the default mib/oid that it is polling and apply only specific mibs that I need.
If I run a snmpwalk for that specific mib to the same device I get a quick reply within microseconds. but in nagios when run cmk -I devicename it takes at least 1mins to 5 mins to poll that device.
The cmk -I hostname is not done regularly in a short time interval. For the normal check of your device only the really needed OIDs are pulled.
A a runtime of 1-5 min for a discovery run is long. The biggest difference is that your manual snmpwalk is not pulling all the enterprise OIDs automatically.
To inspect where the discovery takes so long da a cmk --debug -vvI hostname Now you see at witch point it is slow or waits some time. Write down the OIDs and then you can grep over the folder ~/share/check_mk/checks to search for a check that uses this OID. This check type can be disabled if not needed with a WATO rule.