activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Tim Bain <tb...@alumni.duke.edu>
Subject Re: Active MQ 5.15 acknowledgment issue: 'Async error occurred: javax.jms.JMSException: Unmatched acknowledge - Could not find Message-ID in dispatched-list start of ack'
Date Wed, 24 Jan 2018 13:56:30 GMT
Is this a single broker or a network of brokers? If it's standalone, then
yes, I believe that it indicates a client error. If it's a network of
brokers, we'd need more information about the network (# nodes, topology,
etc.) and about the networkConnectors as you've configured them, to see if
this is a client error as you suggested or if it's an artifact of the
network itself.

Tim

On Jan 17, 2018 2:05 AM, "pablo1a" <patrick.lode@linefeed.nl> wrote:

> Hi all,
>
> We are experiencing a recurring message acknowledgment issue in the form
> of:
>
> /<ActiveMQ Transport: tcp:///127.0.0.1:62894@31313> <<anonymous>> <>
<>
> <1516176117355> <BEA-000000> <Async error occurred:
> <b>javax.jms.JMSException: Unmatched acknowledge*: MessageAck {commandId =
> 17, responseRequired = false, ackType = 2, consumerId =
> ID:c27791-62893-1516176115418-1:1:2:1, firstMessageId =
> ID:c27791-62370-1516175410475-1:1:1:1:1, lastMessageId =
> ID:c27791-62439-1516175599485-1:1:1:24:1, destination =
> queue://Consumer.core.VirtualTopic.FotowebEvents, transactionId =
> XID:[48801,globalId=035763fffffff035ffffffb7ffffffee4d52,branchId=
> 7765626c6f6769632e656a622e636f6e7461696e65722e4a4d53436f6e6e
> 656374696f6e506f6c6c65722e466f746f776562496d6167654576656e74
> cffffffc7ffffffc9ffffffa4],
> messageCount = 2, poisonCause = null}; *Could not find Message-ID
> ID:c27791-62370-1516175410475-1:1:1:1:1 in dispatched-list (start of
> ack)*>
> /
>
> What I see occuring in the broker's log is a two 'Acknowledging message for
> ...' lines occurring only one second apart, for this referenced message ID
> /c27791-62370-1516175410475-1:1:1:1:1/. So while the first acknowledging
> step/call has already dequeued/acked the message, the second, superfluous
> call doesn't find the message anymore thus causing the error.
>
> Questions:
> - Am i right to issume this is erronous behavior *on the client side*, in
> calling the ack twice?
> - What client configuration or setting(s) might be erroneous here?
> - Any other thought for troubleshooting? I have Googled quite a bit and
> applied 'jms.prefetchPolicy.queuePrefetch=1' on the broker URl, but to no
> avail yet.
>
> Context:
> - Consumer is an MDB deployed in weblogic JEE server, set up to use
> ActiveMQ
> 5.15 client libs.
> - XA transaction-based acknowledgement is used.
> - Broker is a ActiveMQ 5.15 broker instance deployed in another weblogic
> server (packaged within war).
>
> Thank you in advance,
>
> Patrick
>
> Broker's log example, red lines mark identical Ack calls happening within
> second from one another:
> <http://activemq.2283324.n4.nabble.com/file/t378877/log-
> extract-unmatched-ack-0.png>
>
> <http://activemq.2283324.n4.nabble.com/file/t378877/log-
> extract-unmatched-ack-1.png>
>
>
>
>
>
> --
> Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-
> f2341805.html
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message