What was the issue with MinConcurrencyTest?
Note that the following change doesn't look correct either:
+ long startTime = System.currentTimeMillis();
+ while (concurrencyReached < expectedConcurrency
+ && System.currentTimeMillis() < (startTime + 5000)) {
+ synchronized(concurrencyReachedLock) {
+ concurrencyReachedLock.wait();
+ }
+ }
If the concurrency doesn't reach the expected level, this will cause the
test to hang (except if at least one message takes more than 5 seconds
before being received).
Andreas
On Fri, Nov 21, 2008 at 18:40, Asankha C. Perera <[EMAIL PROTECTED]> wrote:
> Hi Andreas
>
>> 1. The JMS Javadocs specify the thread safety characteristics of the
>>> Session object as follows:
>>> ...
>>>
>> +1.. I agree my intention was to follow the Spring model, and I will fix
>> this up.. Thanks for catching this :-) !
>>
> This is fixed in the private branch, and I also found the cause for the
> intermittent issue with the MinConcurrency test, which was a threading bug
> in the test. So all tests pass now ..
>
> I think we are all set to commit this to replace the current code now..
> WDYT?
>
>
> asankha
>
> --
> Asankha C. Perera
> http://adroitlogic.org
>
> http://esbmagic.blogspot.com
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
