You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@activemq.apache.org by "Lari Hotari (JIRA)" <ji...@apache.org> on 2008/01/18 18:36:23 UTC
[jira] Issue Comment Edited: (AMQ-1445) RecoveryListenerAdapter -
Message id ID... could not be recovered from the data store!
[ https://issues.apache.org/activemq/browse/AMQ-1445?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=41074#action_41074 ]
lhotari edited comment on AMQ-1445 at 1/18/08 9:34 AM:
-----------------------------------------------------------
I have this same problem with ActiveMQ 5.0.0 version that I downloaded 2 days ago. I don't think that this is fixed in the release build.
The log has only this row:
2008-01-18 17:29:10,769 [/127.0.0.1:2257] ERROR RecoveryListenerAdapter - Message id ID:mycomputer-2250-1200669948525-0:1:30:8:1 could not be recovered from the data store!
ActiveMQ stalls after I do this: send 300 message to a queue, another app replies to each message to another queue. There's a short delay and I after I try again it will always stall after the 1. message. Really strange. Restarting ActiveMQ doesn't help. The data directory has to be deleted.
The messages sent to the queue are non-persistent. I'm using Kaha store.
Both sides are using Jencks + ActiveMQ RA + Spring TransactionTemplate + JdbcTemplate. The receiving side uses Jencks POJO (inbound) + JDBCTemplate (Jencks outbound configuration), inside XA transaction using Spring Transaction configuration for Jencks. I tested with non-persistent and persistent messages and the result is the same.
I also tried Jdbc persistence (derby), but then some messages are lost (EOFException on client before problem, server side says that some acknowledge cannot be matched).
Maybe Jencks is causing problems? Are there test cases that use Jencks + ActiveMQ RA?
was (Author: lhotari):
I have this same problem with ActiveMQ 5.0.0 version that I downloaded 2 days ago. I don't think that this is fixed in the release build.
The log has only this row:
2008-01-18 17:29:10,769 [/127.0.0.1:2257] ERROR RecoveryListenerAdapter - Message id ID:mycomputer-2250-1200669948525-0:1:30:8:1 could not be recovered from the data store!
ActiveMQ stalls after I do this: send 300 message to a queue, another app replies to each message to another queue. There's a short delay and I after I try again it will always stall after the 1. message. Really strange. Restarting ActiveMQ doesn't help. The data directory has to be deleted.
The messages sent to the queue are non-persistent. I'm using Kaha store.
> RecoveryListenerAdapter - Message id ID... could not be recovered from the data store!
> --------------------------------------------------------------------------------------
>
> Key: AMQ-1445
> URL: https://issues.apache.org/activemq/browse/AMQ-1445
> Project: ActiveMQ
> Issue Type: Bug
> Components: Message Store
> Affects Versions: 5.0.0
> Reporter: David Sitsky
> Assignee: Rob Davies
> Priority: Critical
> Fix For: 5.0.0
>
> Attachments: log.txt
>
>
> My application works correctly with JDBC persistence using derby, or indeed with no persistence turned on.
> If I use the AMQ store however, very early on in the application's run, I get the following error, and then my application hangs. This happens for both the VM and TCP transports.
> [VMTransport] 35346 ERROR org.apache.activemq.store.amq.RecoveryListenerAdapter - Message id ID:tamarama-60079-1191564856100-2:2:3:3:4 could not be recovered from the data store!
> I have attached a debug trace of the application, which shows the lifetime of message tamarama-60079-1191564856100-2:2:3:3:4.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.