[
https://issues.apache.org/jira/browse/AMQ-4588?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kevin Earls updated AMQ-4588:
-----------------------------
Attachment: AMQ-4588.patch
Here's a patch that does the following:
- adds JVM parameter activemq.idgenerator.port which can be used to set a port
other than 0
- changed the logging to only print the stack trace if trace logging is enabled.
I did not add a test because of the difficulty of properly unit testing things
that happen in static initializers and logs. I manually tested these cases:
- default, activemq.idgenerator.port not set
- activemq.idgenerator.port set to a valid port
- activemq.idgenerator.port set to invalid values, such as 12345678 or foo
I also changed logging levels and checked for proper output in the logs.
> org.apache.activemq.util.IdGenerator calls non-configurable new
> ServerSocket(0);
> --------------------------------------------------------------------------------
>
> Key: AMQ-4588
> URL: https://issues.apache.org/jira/browse/AMQ-4588
> Project: ActiveMQ
> Issue Type: Bug
> Components: JMS client
> Affects Versions: 5.8.0
> Reporter: Bill DeCoste
> Priority: Minor
> Attachments: AMQ-4588.patch
>
>
> In order to have the AMQ client run on OpenShift the IdGenerator call 'new
> ServerSocket(0);' needs to be configurable to be avoided. OpenShift only
> allows binding to a specific, provided, unique loopback address and a limited
> set of ports. Ideally there would be a configuration option that allows the
> stub prefix to be set via a JVM param. The current logic doesn't break
> OpenShift as the PermissionDenied exception is caught and swallowed but the
> stack trace is ugly.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
