Mon Jul 25, 2011 5:22 pm
Yes this is so frustrating 0 is not down. If you get a result from SNMP and it is 0 that is a VALID result but the dude has been broke for years in this aspect. You can find several posts of mine trying to deal with it. For now I use a function to get the value and add 1 to it then I subtract 1 from it when I check the value. A Cisco CPU can run an average 0%. I suppose I could subtract 1 when I graph as well...
Also to eliminate false positives I reduce the negative cache time to 10 seconds and the cache time to 5 seconds as well as return false instead of 0 if the device is not responding. This takes a lot of extra work but once done the dude works much smoother.
Cisco_CPU function
if(array_size(oid_column("1.3.6.1.4.1.9.2.1.57", 10 ,5)), oid("1.3.6.1.4.1.9.2.1.57.0", 10, 5)+1 ,"False")
Cisco_CPU Probe
Available:Cisco_CPU_a() <> "False"
Error: if(Cisco_CPU_a()<>"False",if(Cisco_CPU_a() -1< 70, "", concatenate("Warning: high CPU = ", Cisco_CPU_a(), "%")), "Cisco Device down")
Value: oid("1.3.6.1.4.1.9.2.1.57.0", 10, 29)