Notification question

[BanditBBS]

I know this has been asked before, and quite frankly I may have been the one asking...but can't find it....

If I have a "DISK CRITICAL - free space: /ua1002 848 MB (4% inode=90%):" and a few days later another disk goes critical and say I have this: "DISK CRITICAL - free space: /ua1002 848 MB (4% inode=90%): /var 100 MB (4% inode=90%):"

is there a way to have nagios be smart enough to detect the message is different and to send out another notification? basically treat it as a new CRITICAL hard state?

[Box293]

Not Nagios directly but I've seen some suggestions that say you could use a wrapper script to export the critical info to a temp file. Then if the critical info in the text file changes you could send a custom notification using the command pipe.

[questrad]

In my opinion each critical alerts must be handled and resolved.
In our company we always trying to have Nagios as green as possible.
If you need to monitor different folder/mount points in one check - I think better not doing it.
Monitor each critical folder(s)/mount point(s) using different checks and always handle the critical - do not leave the critical for even for 1H unresolved (your situation - few days)

[tmcdonald]

I tend to agree with questrad's line of thinking here. I try to never double up on checks unless it makes sense to do so, for example "If 3 of these 5 things are down then alert", but then BPI is good for that sort of thing.