I’m looking at doing some basic SNMP caching for my Nagios system, and wondered if anyone was doing the same and why or why not.
Here is the issue. I am checking about 70 hosts, with a total of about 300 services. Many of the checks use SNMP. This is not an unworkable amount of traffic, but it seems like I could reduce the traffic and the load on the devices being checked if I could cache the SNMP results they return.
I am only looking at a very short caching duration, something like 2 or 3 minutes. Its just that some devices have 5 or 6 checks being run against them over SNMP, so that means 5 or 6 separate queries basically returning the same data to be parsed through.
I’m not sure if the idea of a caching proxy would work, or if there are good reasons against it. What do you guys think?