At 01:48 PM 10/20/2005, you wrote:
However surely the fix is to have the server comms done in a seperate thread

I am working on this - it isn't an easy task but I have a prototype up and
running.

This doesn't solve the underlying problem.  The bug will hang the communication
thread meaning mprime can't contact the server until a reboot occurs.  You'll
probably run out of work before that happens.

or, possibly better still, fork off
a seperate process to do the server comms, which terminates itself when the
job is done.

This is an idea. While it is dangerous to kill a thread, killing a rogue process
ought to be safe.

Is there an
underlying problem with the server which is causing the lockup? Like, server
sets up TCP session but data doesn't transfer for some reason, but TCP
session persists causing problem on retry after timeout?

No matter how badly the server behaves, it should not lock up the client.
I've tried everything I can think of to make the client communication
code bulletproof, but I've failed.  If anyone has any ideas for improving the
robustness of the primenet.c source code, please, please let me know.

_______________________________________________
Prime mailing list
[email protected]
http://hogranch.com/mailman/listinfo/prime

Reply via email to