[Mono-dev] mod_mono and xsp patches

[Joshua Tauberer]

Miguel de Icaza wrote:
>> In my case, I was having problems when the number of concurrent requests 
>> went above 20 or 25. (I forget the details. I've been using this patch 
>> for around 3-4 months... modulo changes I made today.)
> 
> Do you think this is a manifestation of this problem:
> 
> 	http://www.mono-project.com/Article:ThreadPool_Deadlocks

I can't say for sure, and because I was debugging this back around 
September I don't even remember what conclusions I came to then. But I 
remember it certainly felt like a ThreadPool-type deadlock, and it was 
exactly around the right number of requests that ought to cause a 
deadlock if there are (as it seems) 50 threads in a ThreadPool by default.

> If so, it would be interesting to find out who is the consumer of the
> other threads in the thread pool.

Is it (or was it) still the case that xsp has its own ThreadPool? I 
didn't figure out where that was or who was using whose workers. But, 
IIRC, xsp uses one worker per request on the side that communicates with 
mod_mono and a second one per request on the side that starts up the 
ASP.NET routines. Then there are other workers used in ASP.NET 
processing, plus corlib-internal (etc.) workers. So there are quite a 
few per request.

> I wonder if we should set this limit in the managed code.

Not sure it would help, except to set the limit very high when it 
becomes a problem, which we can already do.

-- 
- Josh Tauberer

http://razor.occams.info

"Yields falsehood when preceded by its quotation!  Yields
falsehood when preceded by its quotation!" Achilles to
Tortoise (in "Gödel, Escher, Bach" by Douglas Hofstadter)