Bruce Snyder escribió:
On Thu, Nov 6, 2008 at 11:24 AM, Manuel Teira Paz <[EMAIL PROTECTED]> wrote:
Hello again.
Actually, we need both sides of the socket to get full at nearly the same
time for the deadlock to occur. But the posibility is there.
This is a common problem that has been taken care of in the ActiveMQ
trunk (which will be ActiveMQ 5.2) via the following issue and the
patches attached to it:
https://issues.apache.org/activemq/browse/AMQ-1993
Thanks for the information Bruce. Unfortunately, we have not the
resources or time for a change to that version in the production
environment (I hope we will have in the near future) . I've taking a
look at the information you sent and it looks a great solution for the
problem. In the mean time, perhaps we could try to backport the solution
to the 4.1 codebase to see if it works correctly.
Best regards.
We're currently testing these patches as part of ActiveMQ 5.2 RC4 (the
third release candidate for the 5.2) and voting on it. Please download
it to test it out in your environment and give us your feedback.
Here's the email about the release candidate and where to get it:
http://markmail.org/search/?q=activemq+5.2.0+rc3#query:activemq%205.2.0%20rc3+page:1+mid:6unahrpbm5yegs6i+state:results
Bruce
--
perl -e 'print unpack("u30","D0G)[EMAIL
PROTECTED]&5R\"F)R=6-E+G-N>61E<D\!G;6%I;\"YC;VT*"
);'
Apache ActiveMQ - http://activemq.org/
Apache Camel - http://activemq.org/camel/
Apache ServiceMix - http://servicemix.org/
Blog: http://bruceblog.org/
