Shibboleth Developers

["Scott Cantor" <>]

> Not at the same time no... But if the threads don't clean up their act
> when they finish then eventually problems will occur.

That's true, but I assumed the person that wrote that code knew enough to
clean up if it wasn't automatic. But I admit I can't find any code path so
far that would do that.

The point I was making about the number of sockets was that I couldn't see
how a non-prefork install would generate enough sockets to matter that
quickly, but it's hard to say.
 
Anyway, if I'm reading it correctly, each child thread gets its own SVCXPRT
pointer created in that function. I would suggest storing it in a class
member and then repeating that cleanup code inside the destructor to see
what it does.

My fear, though, is that even if it affects the memory usage, it will create
some cleanup crashes because something gets done twice, so we'll have to
see.
 
> Ok... For this test I used the original test-client. I ran it a couple
> of hundred times and the memory build up.. Then I let it rest for 10
> mins... Then started it up again. As you can see the memory is not
> reused.

Ok. And on Windows, all the web servers are multi-threaded and don't recycle
processes. Now it makes sense to me why it's running differently there.

-- Scott