[
https://issues.apache.org/jira/browse/GEODE-7755?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17029192#comment-17029192
]
ASF subversion and git services commented on GEODE-7755:
--------------------------------------------------------
Commit 44f4c2198c8ec0d7b717c3c940522288b65ab75e in geode's branch
refs/heads/bump/v1.13.0 from Darrel Schneider
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=44f4c21 ]
GEODE-7755: change Pulse to not keep locator alive (#4661)
* Cluster now makes itself a daemon if embedded.
If the JMXDataUpdater loses its connection it now tells the Cluster it is no
longer connected.
* removed debug logging
> Embedded Pulse has non-daemon thread that can keep the process it is embedded
> in alive
> --------------------------------------------------------------------------------------
>
> Key: GEODE-7755
> URL: https://issues.apache.org/jira/browse/GEODE-7755
> Project: Geode
> Issue Type: Bug
> Components: pulse
> Reporter: Darrel Schneider
> Assignee: Darrel Schneider
> Priority: Major
> Fix For: 1.12.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When Pulse is embedded in another service (like the locator) the Thread
> implemented by the org.apache.geode.tools.pulse.internal.data.Cluster class
> can keep the JVM Pulse is embedded in alive. A related problem is that the
> run() method on Cluster will keep running even if it loses its connection
> (which happens if the locator it is embedded in is stopped).
> The combination of these two can cause gfsh stop locator to never see the JVM
> running the locator to exit.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
