As the picture is suggesting, I cannot comprehend how the exception count number can be so different when I am looking at the summary page (the upper portion of my screenshot), and the exception detail page (lower portion)? I removed some sensitive URL and node info from the picture but that's not relevant of course. So: - When I open the errors->exceptions page to find what are the exceptions with the biggest count number, I find that http error 500 appeared 1115 times in the last 2.5 hours. That's 9 exception per minute according to appdynamics, but when you divide 1115 over 150 (minutes) you get 7.43 - which should yield either 7 or 8 exceptions per minute, not 9. But I can live with that, we have something else here obviously, doesn't matter. Fine. - Then I open the details for http 500 exceptions to check it out, only to find that the controller is showing me three URLs with 1 exception, and one URL with 266 exceptions; that is 269 exceptions in total which is faaar awaay from 1115. Not fine at all. - Not only that, but when we look at the diagram on the same details page (see attached picture), we can see that indeed there were around 1.1k counts (roughly 532 times 2 = 1064, and plus few of other data point counts... you get 1115 just fine), but it is extremely different than 269!? I was not lazy and did the same math for other non-http exceptions and the summary page count was always giving the same result as the one on the diagram. Not only that, but when I analyzed other http errors (401 and 404) which were counted in single digits, the number always matched - unlike for the error 500. Anybody has any idea what is going on? This is an obvious difference and it gives me fear that in some cases appdynamics is not giving proper results - which is not acceptable for the premier monitoring solution after all. Please help!
... View more
