I know you would love to just click a couple places and be done well that is not how this is going to go. Typically when you want to monitor a computers hardware SNMP stuff that is not included in the default view you need to add a program that will populate the fields with the data from the hardware you are interested in. With that said double check to see if there is something povided by your hardware vendor.Hello! Can anyone help me with a probe for monitoring S.M.A.R.T. status of HDD? I need this for Windows Server 2008 R2.
In linux you can just add an open source tool to populate HDD Smart data since it is somewhat of a standard http://sourceforge.net/apps/trac/smartmontools/wiki In Windows you should try to get one from your hardware vendor or from the drive manufacturer. Either way once you get the area to be populated then you can write probes against it. Here is a guy testing out the linux SNMP information http://en.doc.centreon.com/HowToUseSNMPAgentExtension
Using SNMpwalk on win2k8 server I could not find any smart entries... Once you have the oid being populated the rest is fairly straight forward.
Here is a template
Create a new function
HDDSmart_test
if(string_size(oid("hddsmartoid", 10 ,5)), oid("hddsmartoid", 10, 5)+1 ,"False")
The above function returns the OID of hddsmart +1 NOTE: if hddsmartoid = 0 when the disk is fine you have to add 1.
Then make a probe of type function that uses the HDDsmart function in the available and error lines (and graph line if needed)
Available: HDDSmart_test()<>"False"
Error: if(HDDSmart_test()<>"False",if(HDDSmart_test()=1,"","Check HDD Smart Status"),"Could not read HDDSmart SNMP info")
Notice I didn't subtract 1 from the Error line I just used the probe as if it was 1 when true so that I didn't have to try to test for 0 which is always false. The dude "if" statement always return false on SNMP values of "0" even though this is a valid return from a probe and this "feature" makes building accurate probes complicated.
HTH,
Lebowski