[
https://issues.apache.org/jira/browse/QPIDJMS-186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robbie Gemmell closed QPIDJMS-186.
----------------------------------
Resolution: Invalid
Please use the users mailing list for questions such as this, JIRAs can be
raised upon discovering actual bugs or improvements needing changes to be made.
I see you did actually post to the dev list too at the same time: please dont
post identical questions to both places at the same time, it justs gives two
different places for us to reply and update. I'm commenting here as I close
this JIRA, because the rest of the discussion is here already.
The 'about 100 seconds' may related to idle-timeout handling, though its
unclear due to the general lack of information in your post. The newer client
(0.9.0) defaults to enforcing a 2 minute idle-timeout. The old 0.32 Java broker
didn't support sending heartbeats to satisfy that requirement, so the client
could think the connection is dead if it doesnt receive any traffic from the
broker for that length of time. You could try configuring the clients idle
timeout config to 0 to disable it:
http://qpid.apache.org/releases/qpid-jms-0.9.0/docs/index.html#amqp-configuration-options
to see if that helps. Its unclear from your description if that is actually in
play or even if it is the only issue in play (I'd guess there may be others,
such as an equivalent of:
http://mail-archives.apache.org/mod_mbox/qpid-users/201606.mbox/ajax/%3CCAFitrpQkSbXWee_k24LPx%3DCVXne20EYS-z3kdogoATFvVHW23A%40mail.gmail.com%3E).
You might want to consider turning up the logging and or protocol-tracing
(http://qpid.apache.org/releases/qpid-jms-0.9.0/docs/index.html#logging) to see
if more detail is available, though given the next bit there may be no point.
In any case it appears that you are saying things work fine with the
newer/current 6.0.3 broker release, which will likely be because the newer
broker releases have bug fixes and features (such as heartbeating) that arent
available in older versions, which you should upgrade to take advantage of.
> [Borken-0.32,qpid-jms-client0.9.0] When recieve message, after about 100s,
> throw error" The MessageConsumer was closed due to an unrecoverable error"
> -----------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: QPIDJMS-186
> URL: https://issues.apache.org/jira/browse/QPIDJMS-186
> Project: Qpid JMS
> Issue Type: Bug
> Components: qpid-jms-client
> Reporter: zjiali
> Priority: Blocker
>
> I using "MessageConsumer.receive ()" to receive message.
> case 1: when I using (Borken-0.32,qpid-jms-client0.9.0), after about 100
> seconds, throw error" The MessageConsumer was closed due to an unrecoverable
> error".
> case 2: Based on case 1, I changed to connect Borken-6.0.3 server, it works
> without any error.
> case 3: Based on case 1, I changed "qpid-jms-client0.9.0" jars as
> "qpid-amqp-1-0-client-jms-0.32" jars, and connect with Borken-0.32, it works
> without any error.
> qpid-amqp-1-0-client-jms-0.32" jars download from:
> https://qpid.apache.org/releases/qpid-0.32/index.html
> so my question is why qpid-jms-client0.9.0 + Borken-0.32 can not work?
> I searched out a similar issue of ActiveMQ.
> http://qpid.2158936.n2.nabble.com/Qpid-JMS-0-9-0-receive-hanging-issue-relates-to-Artemis-works-on-ActiveMQ-td7642585.html
> Regards
> -Gary
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
