Call for testing: Fixing a race condition in MATPE

[Ric Klaren]

Hi,

On 13 January 2011 07:27, "이희승 (Trustin Lee)" <trustin at gmail.com> wrote:
> According to the recent reports, (Ordered)MemoryAwareThreadPoolExecutor
> has a race condition that results in an intermittent I/O freeze.
>
> I have committed the fix to the repository and tested it fairly, but I
> want to make sure if the fix does really fix the race condition.  If you
> were affected by this issue, please download the latest snapshot JAR and
> let me know if the problem is really gone (or still there):
>
>  http://gleamynode.net/files/netty-3.2.4.Final-SNAPSHOT.jar

Is there already a release planned of this version (or a indication of
a time frame?), as there seem encouraging reports with respect to the
race condition?

I'm asking since we recently rewrote a netty using component to use
OrderedMemoryAwareThreadPoolExecutor and I'd rather ship our release
with a netty that does not have a known issue like this.

If I have some idea of the timeframe then we can decide what to do.
Vendor branch or ship with the current version and push a bundle
update once the new version arrives.

Thanks in advance,

Ric