I have a server running Apache which is proxying requests to Tomcat for an application I'm running. I was able to monitor an IpService for "http" on port 80 and that appears to be working fine.
The application I'm running on Tomcat has a status page with the message "ok". I want to run a separate check to GET that page and use the expected regex to search for "^ok$". I was able to accomplish this by adding tcp_00080 to the service keys box on the new IpService (although it's not documented that this is required). One oddity that I've noticed is that if I go back now and edit the http check, it automatically changes the service class to the custom one!!!! Anyway, I purposely disabled the Apache proxy to Tomcat as a test, yet the status still reports green. I have confirmed via the access logs that the URL is being requested by my Zenoss box: xxx.xxx.xxx.xxx - - [09/Jul/2009:11:23:51 -0400] "GET /idp/profile/Status" 404 216 "-" "-" The reason that I want them to be separate is so that I can know if there was a misconfiguration in the application that would not be picked up via a simple TCP port check. An alert at 4am specifying "http is down" or "x_application is down" could save time and sleep because it would be fairly obvious what needs to be looked at. Does anyone have any ideas why this might not be working? I've tried remodeling the device to no avail... the regex appears not to be running. We just updated to 2.4.2 within the last hour and it's still happening. I'm pretty much out of ideas at this point. -------------------- m2f -------------------- Read this topic online here: http://forums.zenoss.com/viewtopic.php?p=36830#36830 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
