Tavin Cole wrote:

 > Nevertheless, coding the thread pool to resist bugs in the rest of the
 > node (and adding a lot of complication to it in the process) seems like
 > the wrong approach in the long run.

You do not like fault tolerant systems !?!

If freenet fails, I do not want to reboot my machine (which I had to two 
times,
because it did not respond anymore).

Also it is perfectly ok (imho) to have 200 jobs waiting to be
executed, but it's no good running them all at the same time.

The rest of the code should care about not accepting too much work in
the longer term, but should not have to care about short term issues.

 > Switching to a thread pool that doesn't create overflow threads won't
 > solve this.  It'll just make the node die sooner when the ticker can't
 > get any more threads.

Actually it does. The node does not respond for some time, but after
that it gets back to normal (there might be some dropped connections),
which is much harder, if it accepts much more work, before stopping to
respond.

Niklas


_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://lists.freenetproject.org/mailman/listinfo/devl

Reply via email to