Hi,
To confirm with Chris, I've only been able to run the sample test
program I posted on 2 and 4-processor machines.  I still have yet to try
it on a single processor machine.  I suspect that on multi-processor
machine, the multi-thread issue with EMPTY_DOC will be more apparent.
The threads will be working more truly in parallel, whereas they'd be
running almost serially on a single-processor machine since they have to
all share the same CPU time.

If anyone else has similar findings, please post to let us know:)
-Vinh


-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
Sent: Thursday, September 13, 2007 3:20 AM
To: [email protected]
Subject: RE: RE: EMPTY_DOC thread stability issues

Hiya,

Vinh uploaded a test case that fails on MT machines only.  It would be
great, if you have access to an SMP or multicore box, to try the tests
out and see if you can replicate the lost messages.  It may be unrelated
to the EMPTY_DOC issue but as I can't replicate it I can't help in
solving it :-<.

cheers,
Chris

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Tuesday, September 11, 2007 10:06 PM
To: [email protected]
Subject: Re: RE: EMPTY_DOC thread stability issues

Who is working on a comprehensive patch for the EMPTY_DOC issue
(MUSE-270) and when will it be available?  I'm willing to help if
needed.  Thanks!


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to