This is the case:
We have applications installed on customers servers which when error occurs email error logs to us. The application server is unreachable from our network and therefore we can
the idea which comes to mind (there is a better solution for sure) is to script the analysis of the emails. when the email is analysed give an output depending on the email you received, OK, WARTNING or CRITICAL then move the mail message to some other directory, (if there is no inbound email it should be CRITICAL). Then create an snmp action which runs the script you created and call a check_snmp from nagios to the localhost.
Luca
Does the remote host have to ability to ping the nagios server? If not, can the remote host run nsca and some plugins and deposit those results in some folder such as an ftp site? Surely they could do that. This result of the active service check could then be grabbed by the central nagios server through a cronjob that ftp’s to the site, gets the file, and stuffs it in the var/rw/nagios.cmd file.
You said the remote application emails error logs to you. So just use “check_log” nagios check plugin to look at this log for the events you are interested in.