Today’s challenge — Monitor and graph the air conditioning system to track performance.
With an uptime server installation already in place, I set about adding snmp v2 monitoring to track things like air temp, status, etc. By default, the uptime monitoring software doesn’t have built in support for my air conditioning system (apc inrow sc 100) so I had to add it in manually using the below steps:
I did a quick search around the web for any existing plugins/support for what I was attempting to do, and found a post using cacti from here. Unfortunately the cacti plugin wasn’t supported on uptime, nor was it for the correct apc model, but it did get me started on the correct MIB information and what I should look to track/graph.
These SNMP were made for walkin’:
With my end goal clearly possible over snmp v2 I set out to find the exact OIDs for my specific Inrow device. I grabbed a windows tool to allow browsing MIB files and to perform snmpwalk so I could see just what information was available to me.
Success! using the system defaults I was able to get some good information out of the Inrow over snmp, but what does sysServices.0 mean… and where are all the other cooling/temp variables I was promised? Time to pull the apc MIB file from their documentation and upload it to my snmpwalk application.
MIB browsing and more:
After loading up the APC PowerNet MIB into ManageEngine I came across an all to common discovery, way too many settings and options! How was I supposed to know where the OID’s I needed were in the hundreds of options for the APC MIB? GETNEXT to the rescue! By performing GETNEXT on the main entry of the APC MIB I was able to find what section my APC Inrow would respond to, putting me in the ballpark for some manual searching.
As seen above, I found the folder containing all of the relevent status information for my particular Inrow! With this, including the direct OID’s in hand, I headed over to the uptime software to start adding tracking information. The hard part was over.
Uptime settings and wrapping up:
Hard to believe but we’re almost done! The final steps involve dealing with uptime and how it listens to snmp. I punched my snmp information into the uptime system, added my inrow device to be tracked, and then set to work plugging in the OID’s I wanted to track.
Manually adding the name/OID/description was tedious, but after coming this far it wasn’t so bad. I manually gave uptime the OIDs to track, let it know what data type each was, and gave them some basic descriptions so the charts would be easy to make sense of. Saving and then running a test of the new snmp OIDs being tracked showed success!
Flip a few switches in uptime, and I had email alerts, pretty graphs, and a full array of information streaming from my Inrow A/C to my uptime monitoring software. With this new skillset in hand you can snmpwalk across MIBs and collect OIDs for just about anything, adding any number of snmp enabled devices to your monitoring network, good luck!