In my test environment we have servers with many apps. Each with it's own app pool. These apps typically share similar libraries and namespaces. I would like to be able to segregate custom profiling by app pool so we can turn up custom profiling for a specific app pool but not others.
Example:
In ProfileClientConfig.txt
[
{
"appPool": "app_pool_1",
"Class": "Some.Namespace.Class",
"Method": "*"
},
{
"appPool": "app_pool_2",
"Class": "Some.Namespace.Class",
"Method": "SomeSpecificMethod"
}
]
With this enhancement I would be able to selectively adjust custom profiling per app pool without effecting other app pools. It's worth noting that on occassion, even if only for half an hour, we need to bump up profiling to use "*" to get a good idea of where we need to dig further to identify a performance issue. On servers with many apps this causes CPU to spike. The goal of this enhancement is to allow me to really crank up the profiling for a short time without effecting every app on the server and hopefully prevent CPU from spiking.
See also my comment on https://ideas.stackify.com/ideas/RM-I-66
Another method would be useful would be an option to add somethng like Stacify.ApmExclude into the web.config - this would easily allow some apps to be monitored and others not. This can then be set on deployment
Additionally, they should provide a UI on the website to edit custom profiling. Click on application - then add custom profiling. Be great to be in a trace of some web call - and just be able to click 'custom profiling' on that and have it activate detailed profiling whenever that method is called. (Essentially adding all the information for you). Hand editing a random text file on your server for profiling options is a bit obscure!