[
https://issues.apache.org/jira/browse/OOZIE-1057?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497931#comment-13497931
]
jiezhou commented on OOZIE-1057:
--------------------------------
browsing the source code,I find that if
oozie.service.ActionCheckerService.action.check.interval is set,the
ActionCheckerService.java's init() will read this value.
but ActionCheckXCommand will log a message at another interval from its
executor,the source code is as following:
-------------------------ActionXCommand.java 99th
line---------------------------
long retryDelayMillis = executor.getRetryInterval() * 1000;
action.setPendingAge(new Date(System.currentTimeMillis() +
retryDelayMillis));
LOG.info("Next Retry, Attempt Number [{0}] in [{1}] milliseconds",
actionRetryCount + 1 , retryDelayMillis);
---------------------------------------------------------------------------------
obviously,the executor's interval determine the log message , and the
executor's default value
is 6000 , and the ActionCheckXCommand doesn't call its executor's
setRetryInterval() to control the log message's interval.
> Log message for retrying to connect to the JT always says 60,000 milliseconds
> -----------------------------------------------------------------------------
>
> Key: OOZIE-1057
> URL: https://issues.apache.org/jira/browse/OOZIE-1057
> Project: Oozie
> Issue Type: Bug
> Affects Versions: trunk
> Reporter: Robert Kanter
> Priority: Minor
> Fix For: trunk
>
>
> When Oozie tries to reconnect to the JT, the ActionCheckXCommand will log a
> message like this:
> {code}
> 2012-11-08 16:31:48,153 INFO ActionCheckXCommand:539 - USER[oozie] GROUP[-]
> TOKEN[] APP[map-reduce-wf] JOB[0000002-121108161824321-oozie-oozi-W]
> ACTION[0000002-121108161824321-oozie-oozi-W@mr-node] Next Retry, Attempt
> Number [1] in [60,000] milliseconds{code}
> If you set {{oozie.service.ActionCheckerService.action.check.interval}} to
> something other than 60, it will behave correctly (that is, it will check
> every X seconds instead of 60 and even log the message at that interval), but
> will still print out [60,000] milliseconds.
--
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
