activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kevin Quick (JIRA)" <>
Subject [jira] Commented: (AMQCPP-349) broker restarts undetectable via tcp: transport
Date Tue, 25 Jan 2011 19:31:44 GMT


Kevin Quick commented on AMQCPP-349:

Yes.  The receive() method is being called on a Consumer.  The Consumer was created for the
original broker instance.  The receive() is being called with a 1.5 second timeout (to allow
periodic local processing).  When the original broker is stopped, there does not appear to
be any interruption of the current receive() call, and all subsequent receive() calls return
normally (with a timeout after their 1.5 second period) as well with a null Message pointer.
 It's unclear how many (if any) receive calls were made while the broker was down, but at
no point during the broker restart is there any perturbation or variation in the 1.5-second
receive() operations.

> broker restarts undetectable via tcp: transport
> -----------------------------------------------
>                 Key: AMQCPP-349
>                 URL:
>             Project: ActiveMQ C++ Client
>          Issue Type: Bug
>    Affects Versions: 3.2.4
>            Reporter: Kevin Quick
>            Assignee: Timothy Bish
> A client connected to a queue on a broker via a tcp: transport connection receives no
indication of a broker restart.
> This leaves the client waiting forever for messages on the queue but the broker does
not have a subscription for that client.  The client has no indication that it must initiate
a reconnection with the broker.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message