Everything is working great here, but for some reason I can't monitor the hard disks on any of my linux hosts. For 2 hosts it recognizes the / partition, but doesn't seem to get any information in the graphs. I played with it quite a bit and can't seem to get it to work properly.
Thanks for you time,
NetMRG's disk graphing uses the UCD disk MIB (as opposed to the Host disk MIB which is sometimes used). The snmpd.conf on the monitored host needs go have lines like this for each partition
When this is done, you should be able to get information from snmpwalk
[code1906acc4c96][balleman@pine balleman]$ snmpwalk -v1 -cpublic oak dskPath
UCD-SNMP-MIB::dskPath.1 = STRING: /
UCD-SNMP-MIB::dskPath.2 = STRING: /storage[/code1906acc4c96]
This information (and a bit more) should be present in [b906acc4c96]View Disk Cache[/b906acc4c96] (on the Devices page). If the info there isn't correct or complete, or the link is disabled (meaning the cache is empty), click [b906acc4c96]Recache Disks[/b906acc4c96] to have NetMRG poll them again.
The current NetMRG doesn't have a "Monitor/Graph Disk" option as it does for network interfaces (laziness on my part
). For now, you'll have to do it manually. For each disk you want to monitor, create a new subdevice ("disk" type). For indexing purposes, you'll need to add a parameter to this subdevice. A parameter with name "dskPath" and value "/storage" will monitor a partition mounted on "/storage" but you can also use "dskDevice" if you know the device (i.e. "/dev/hda1") or "dskIndex" if you want to match it with a fixed index. After the parameter has been added, you can add the monitors to the subdevice (Disk - Used and Disk - Available are what I normally use).
If you've already done all of that, we can look over the output of the gatherer to see what's wrong. Please let us know if you get it working or have more problems.
Thanks for the help. I was able to get that to work no problem, however when I make a custom graph I am able to change the multiplier to 1000 so it shows MB instead of kb, but I don't see a place to change this for the mini graphs. It's not really a big deal for me, but for other people they may think it's lying to them ) Keep up the good work folks!
thanks again for all the help,
Yeah, I forgot to mention that "gotcha." Seems strange to me that they'd pick a non-base unit for reporting disk sizes, but the Host MIB is worse (uses block counts, I believe).
Of course, these issues are trivial compared to the DS1-MIB which has nothing in it which can be graphed as a gauge or counter with any meaning... thanks AT&T!
Anyhow, I'm glad you've got it working. Disk graphing should be as 'one-click' easy as interfaces in pre2 (mid- to late-month, hopefully). Thanks for helping us get the bugs worked out!