[jbosstools-issues] [JBoss JIRA] Resolved: (JBIDE-9354) Stopping JBoss 5.1 server Fails in about 3 different ways!

[Andre Dietisheim (JIRA)]

     [ https://issues.jboss.org/browse/JBIDE-9354?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Andre Dietisheim resolved JBIDE-9354.
-------------------------------------

    Resolution: Done


code is committed to TRUNK (M3)

> Stopping JBoss 5.1 server Fails in about 3 different ways!
> ----------------------------------------------------------
>
>                 Key: JBIDE-9354
>                 URL: https://issues.jboss.org/browse/JBIDE-9354
>             Project: Tools (JBoss Tools)
>          Issue Type: Bug
>          Components: JBossAS
>    Affects Versions: 3.3.0.M3
>            Reporter: Rob Stryker
>            Assignee: Andre Dietisheim
>            Priority: Critical
>             Fix For: 3.3.0.M3
>
>
> Stopping a JBoss 5.1 server comes up with the dialog introduced in JBIDE-9243. This dialog was meant to give users the option of what to do in the case of a server STARTUP and the server was already up. This should not be showing during a SHUTDOWN. 
> The reason it shows during the shutdown is because StopLaunchConfiguration, a class marked deprecated and unused, is in fact used. This class, StopLaunchConfiguration, EXTENDS a class named AbstractJBossStartLaunchConfiguration, including the new methods for what to do during STARTUP in the even the server is ALREADY STARTED
> Additionally, after correcting the preLaunchCheck for StopLaunchConfiguration, an attempt to stop the server was *always* receiving a failing dialog, even though the server would eventually shut down. The reason for this is that the stop code in LocalJBossBehaviorDelegate.gracefullStop$Thread.run included the following code:
> if (stopProcess.getExitValue() == 0) {
>   if( isProcessRunning()) {
>       // cancel the stop, say it failed, etc
>   }
> }
> This code will *ALWAYS* misread a stop as a fail because the process that is launched to initiate the stop will finish in under 1 second, while the actual SERVER process may take 5 or 6 seconds to shut down. Therefore, after the stop process is launched, and finishes, with an exit code of 0 (no error), the process is ALWAYS still running. 
> I am correcting the following:
>   1) Overriding the preLaunchCheck in StopLaunchCOnfiguration
>   2) I am replacing the faulty logic for recognizing a failed stop with proper logic to ONLY respond if the exit code is NOT 0
>   3) I am expecting Andre to PROPERLY mark as deprecated or NOT deprecated the class StopLaunchConfiguration. It is currently marked deprecated, but it is still in use, and inherits from AbstractJBossStartLaunchConfiguration. A review will be necessary to see if it makes sense for StopLaunchConfiguration to inherit from a class named AbstractJBossStartLaunchConfiguration. 
> And finally, changes that seem like they make no difference and are just code cleanup / re-organization *STILL* require SMOKE TESTS before commits. This is to make sure you didn't accidentally break something, like startup and shutdown of every server type. A bad build hinders QE's work for a full day and can cause tons of other problems. 

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira