[jira] [Resolved] (GEODE-2125) Network partition forceDisconnect leaves crippled Locator process running

["Bruce Schuchardt (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/GEODE-2125?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Bruce Schuchardt resolved GEODE-2125.
-------------------------------------
    Resolution: Not A Problem

The logs clearly show that auto-reconnect went into effect when the locator declared a loss of quorum.  When auto-reconnect is enabled (and it is enabled by default) this is the expected behavior.  Turning off network partition detection, as suggested to the user, will clear this up because the locator will not shut down and go into auto-reconnect.

> Network partition forceDisconnect leaves crippled Locator process running
> -------------------------------------------------------------------------
>
>                 Key: GEODE-2125
>                 URL: https://issues.apache.org/jira/browse/GEODE-2125
>             Project: Geode
>          Issue Type: Bug
>          Components: management, membership
>    Affects Versions: 1.0.0-incubating
>            Reporter: Kirk Lund
>         Attachments: locator_failure-logs.txt, thread_dump.txt
>
>
> If the Locator is started from GFSH and the cluster has one server which is killed, network partition detection will initiate forceDisconnect which shuts down the Locator. The Locator process, however, continues to run and any attempts to use stop locator or status locator will fail.
> The Locator log file is attached as locator_failure-logs.txt
> The Locator thread dump (via jstack) AFTER it has shut down due to forceDisconnect is attached as thread_dump.txt



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