[arquillian-issues] [JBoss JIRA] (ARQ-709) Arquillian doesn't wait for the process to really end, causes problems like "port in use".

Ondrej Zizka (Commented) (JIRA) jira-events at lists.jboss.org
Thu Dec 22 00:27:09 EST 2011


    [ https://issues.jboss.org/browse/ARQ-709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12652731#comment-12652731 ] 

Ondrej Zizka commented on ARQ-709:
----------------------------------

(06:26:26) ozizka: dmlloyd: Is it specific for debug mode?
(06:26:54) ozizka: dmlloyd: Ever seen with eg mgmt port?
(06:27:22) dmlloyd: ozizka: I don't think so; it's just that due to the timing with JDWP (the socket is bound instantly on VM start and kept until VM stop) it's a better detector for the problem
(06:27:46) ozizka: I see
(06:28:08) ozizka: dmlloyd:  https://hudson.qa.jboss.com/hudson/job/AS7-TS-valid/14/testReport/org.jboss.as.test.integration.domain/DefaultConfigSmokeTestCase/test/
(06:28:21) ozizka: 0:08:28,544 ERROR [org.jboss.msc.service.fail] (MSC service thread 1-4) MSC00001: Failed to start service jboss.remoting.server.management: org.jboss.msc.service.StartException in service jboss.remoting.server.management: Address already in use /127.0.0.1:9999 
(06:28:38) dmlloyd: yeah I bet that's the same problem
(06:28:42) ozizka: OKay
(06:28:47) dmlloyd: a busy server could be more susceptible
(06:28:51) ozizka: +1
                
> Arquillian doesn't wait for the process to really end, causes problems like "port in use".
> ------------------------------------------------------------------------------------------
>
>                 Key: ARQ-709
>                 URL: https://issues.jboss.org/browse/ARQ-709
>             Project: Arquillian
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: Base Implementation
>    Affects Versions: 1.0.0.CR7
>            Reporter: Ondrej Zizka
>            Priority: Critical
>
> (05:54:09) dmlloyd: when running with JDWP enabled on the client or server, I occasionally get bind exceptions due to address in use
> (05:54:25) dmlloyd: which means that tests are running into each other without waiting for termination of the previous one
> (05:54:35) dmlloyd: which may also be causing other issues
> (05:56:04) ozizka: dmlloyd: Yes, lbarrerio observed similar problem too,
> (05:56:47) ozizka: And that's arq's issue too - there's no way to get around this currently AFAIK. Or is there?
> (05:57:05) ozizka: Perhaps "manually" wait in @AfterClass or such
> (05:57:07) dmlloyd: yeah, it can wait for the child process to terminate
> (05:57:07) ozizka: which is ugly
> (05:57:14) dmlloyd: I mean arq should
> (05:59:17) ozizka: dmlloyd:  Do you have it somewhere on hudson?
> (05:59:24) ozizka: dmlloyd: It never happened to me actually
> (05:59:49) ozizka: Send me a log if you have one handy
> (06:01:35) dmlloyd: ozizka: no, try running with this command though: 
> {code}
> mvn -DallTests install -Djpda -Dsurefire.jpda.args=-Xrunjdwp:transport=dt_socket,address=8787,server=y,suspend=n \
>                    -Dmaven.surefire.debug="-Xdebug -Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=5005"
> {code}
> (06:13:56) ozizka: dmlloyd: That's on linux?
> (06:14:03) dmlloyd: yes

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        


More information about the arquillian-issues mailing list