incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Hammant <Paul_Hamm...@yahoo.com>
Subject Re: [altrmi] Bug in AbstractCompleteSocketStreamServer
Date Sat, 15 Mar 2003 22:55:09 GMT
Peter,

Ooooh, been looking for something like that. Marathon ( 
http://marathonman.sourceforge.net/ ) is being bugged by an issue that 
sounds like it be caused by that.

Lets go for START & STOP. The intermediate states were just just 
over-design anyway.

Clearly a case of a missing or malfunctionig unit test.....

- Paul

> I found a bug in the AbstractCompleteSocketStreamServer.
>
> The code sets the internal state to STARTING, then fires off the 
> thread to monitor the socket, then sets the state to STARTED.
>
> The thread that monitors the socket does its thing as long as the 
> state is STARTED. I'm seeing the monitor thread start and immediately 
> exit since the state has not yet been changed to started.
>
> My question is, what is the proper fix, skipping the STARTING state 
> and going straight to STARTED? Or set STARTED right before spawning 
> the monitor thread, or in the monitor thread go into a loop if the 
> state is STARTING?
> -pete
>



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message