commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Siegfried Goeschl (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (EXEC-71) Changes for EXEC-34 (Race condition prevent watchdog working using ExecuteStreamHandler) hang endlessly if the command cannot be found
Date Tue, 05 Jan 2016 14:34:39 GMT

    [ https://issues.apache.org/jira/browse/EXEC-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15083140#comment-15083140
] 

Siegfried Goeschl commented on EXEC-71:
---------------------------------------

Wrote a patch where DefaultExecutor notifies the WathcDog that there will never anything to
watch

> Changes for EXEC-34 (Race condition prevent watchdog working using ExecuteStreamHandler)
hang endlessly if the command cannot be found
> --------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: EXEC-71
>                 URL: https://issues.apache.org/jira/browse/EXEC-71
>             Project: Commons Exec
>          Issue Type: Bug
>    Affects Versions: 1.2
>            Reporter: Dominik Stadler
>            Assignee: Siegfried Goeschl
>
> The following test hangs endlessly for me because the ExecuteWatchdog will never get
"processStarted = true" because starting an non-existing process will never lead to the started
flag.
> {code}
> @Test
> 	public void testCommonsExecHang() throws Exception {
>         CommandLine command = new CommandLine("someunexistingcommand");
>         DefaultExecutor executor = new DefaultExecutor();
>         ExecuteWatchdog watchdog = new ExecuteWatchdog(ExecuteWatchdog.INFINITE_TIMEOUT);
>         executor.setWatchdog(watchdog);
>         DefaultExecuteResultHandler resultHandler = new DefaultExecuteResultHandler()
{
> 			@Override
> 			public void onProcessFailed(ExecuteException e) {
> 		        System.out.println("Process did not stop gracefully, had exception '" + e.getMessage()
+ "' while executing process");
> 				super.onProcessFailed(e);
> 			}
>         };
>         executor.execute(command, null, resultHandler);
>         // this hangs!!
>         watchdog.isWatching();
>         watchdog.destroyProcess();
> 	}
> {code}
> not calling isWatching(), but only destroyProcess() hangs as well!
> Options that I see: timeout the loop or handle the case of failure to start the Watchdog
in DefaultExecutor.executeInternal() so that the Watchdog does not hang in this case



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message