In my humble opinion the logs need some rework and are at the moment almost useless when live debugging an environment. The new style logs looks good but they have a major drawback because they lag in time and only give a small timeframe. So when doing live debugging they are useless because either the error has not yet appeared in the logs or if you wait to long they already gone again. So in this caser I always use the old logs. But the old logs take ages to download from the V4 cloud. You sometimes have to wait minutes before receiving the logs. And these logs are cluttered with all the retrieve actions done by the application. This should be a log setting and should not be included as standard in the logs in my opinion. Now the logs take up a lot of space and you need to scroll a lot to find the parts you are looking for. So live debugging is at the moment just a pain in the ....
I already filled a ticket for this but at the moment this is not getting much priority within Mendix.
Probably not what you are looking for, but I wanted to point out that you can still download the log locally and analyze it in any way you want.
Hope this helps,
The new logs are indeed a pain, because as Ronald points out, they dumped in all the requests with the application logs and the viewer (which actually has very handy filter options) only shows you a small time span.
I have a little tip when you're dealing with those downloaded logs though. The lines with application log messages will be marked with a tag like '[APP/PROC/WEB/0]', so if you filter the file for those lines you can get rid of most junk.
(Which is very easy if you're using a command line/terminal:
On a unix terminal, just pipe the output to grep, e.g.: less 'yourlogfile.txt' | grep "APP"
On the windows command prompt, do the same with findstr, e.g.: type 'yourlogfile.txt' | findstr /C:"APP" )