You are doing nothing wrong. This is a common question and misunderstanding. I’ve also run into it. The one-minute-interval and the rule “Normal check interval for service checks” to change it is for active checks only. Simply speaking, the active checks are those that actively fetch data from a host, such as calling the checkmk agent or running some SNMP request against a host.
With this rule you can change the interval of how often an SNMP request is made or the checkmk agent is called.
But, for example, the checks that calculate the CPU or memory usage based on the data from the agents are simply run whenever the agent is called and returns data. You cannot change the interval of the CPU check individually. This would only be possible if there was an active and dedicated CPU check that fetches the data from the device on its own and does not rely on the data that was fetched by some other agent.
Of course you can change the interval for the agent (be it the regular agent or the SNMP agent) but then all other services that rely on that agent data are also affected.
It’s too bad because I’m managing network devices and some of them have more than 1k services which have different severity. I was previously using Zabbix which allows to modify interval for each service.
Anyways, I’ll set a 1min intervall for all critical equipment and see if they is support the load.
As this is SNMP you can to exactly what you want I think. Have not tested it but if you know what sections in OID’s I think you can make a separate call for that using Fetch intervals for SNMP sections
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.