how to detect even a fraction of downtime for the URL

[Satyam]

Hi,

I have a very critical application (URL) hosted on a windows production server on the Jboss service. As now I am monitoring the URL availability and the production server availability using NagiosXI. But as I am polling to check these on an interval of 5 mints and also there is check latency often on my Nagios server because of high load, I am not able to detect even a seconds of downtime of this url/server or reboot of server/restart of jboss service which will effect availability of this URL.

I have pointed out some of the feasible solutions with my concerns.

• Polling interval to monitor the URL and the Server can be set in seconds. (Active Check - Concern is the check latency which is often high, because of my nagios server load)
• Event log monitoring for the Jboss service through windows event log monitoring agent, as document attached. (Passive Check)
• SNMP trap service can be set on the production windows server to get quickly notified of its reboot/down status. (Passive Check)
• Monitoring log files on windows server, as document attached. – Active Checks Concern is the dealy in check, which will not server my purpose.

Please provide your suggestions on the same.

[yancy]

Another possible solution would be to send a passive check to NagiosXI when the jboss service was restarted. If you have a script that restarts the jboss server, post data could be sent to Nagios by the restart script.