UEM - Data - By Browser version
Unlike Synthetic monitoring, reporting on the UEM data is a bit challenging. See the case of the home page data from past few hours. The newer version browsers are reporting slower times than the older ones...
Any suggestions?
Carlos - Sorry for the delay in response. I had opened a case and then we had to go to 6.1 as there was another dependency. I was revisiting this and still see similar behavior. It looks to me more than a compatibility case. I have been asked to enable logging on the visit etc. Will keep you posted.
Hi Jibi, yes that is the expected behavior with compatibility mode.
What you can do is dig a little deeper with Fiddler, i.e. using IE10 and IE 8, to find out what is causing the delay on client side (since network and server are in ms).
If you have a PurePath we can check on (one PP for IE8, IE9 and IE10, I'll be glad to do so, it sounds interesting indeed.
Bests,
Carlos.
Carlos - In my case the issue is more the the newer browsers for example IE10. I believe if IE 10 was in compatibilty mode as IE9, dynaTrace should be picking the useragent as that of IE9 and showing data under IE9.
Yet to dig deeper on the PPs and Compatibilty. Network and server contribution is in ms so they cannot be a factor I am thinking.
Hi,
You may also want to check IF the browsers are or are not forcing compatibility with other/older versions, i.e. IE 9 may be "presenting" the user agent as IE 7 or 8...that is a bit misleading...once you confirm that, you can have a clearer picture...just my 2 cents.
Also, you can check if the "extra-time or time discrepancies" when comparing versions, happens at the browser, client side, network or server, that helps to figure out the actual performance driven by browsers.
I did. See this behavior on other pages too. Opened a ticket for now.
Have you drilled down to some PurePaths to see/compare the transactions? All of this charting data is a graphical representation of the underlying transactions of real users. Drilling down and analyzing the transactions should be the next step to determine what is causing the slow downs.
Tom
