[jira] [Commented] (GEODE-863) Gfsh-started locator exits after a successful auto-reconnect

["ASF subversion and git services (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/GEODE-863?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15119709#comment-15119709 ] 

ASF subversion and git services commented on GEODE-863:
-------------------------------------------------------

Commit 4e211e2c2f771a4c399ce2be8dd0eb85f9b63740 in incubator-geode's branch refs/heads/develop from [~bschuchardt]
[ https://git-wip-us.apache.org/repos/asf?p=incubator-geode.git;h=4e211e2 ]

GEODE-863: Gfsh-started locator exits after a successful auto-reconnect

Modified InternalLocator.waitToStop() to work with auto-reconnect.
Modified InternalDistributedSystem to keep log appenders open during
auto-reconnect.
Added test of waitToStop to ReconnectDUnitTest.
Modified LocatorLauncher to show error details.


> Gfsh-started locator exits after a successful auto-reconnect
> ------------------------------------------------------------
>
>                 Key: GEODE-863
>                 URL: https://issues.apache.org/jira/browse/GEODE-863
>             Project: Geode
>          Issue Type: Bug
>          Components: membership
>            Reporter: Bruce Schuchardt
>            Assignee: Bruce Schuchardt
>
> Gfsh uses the method InternalLocator.waitToStop() to wait for a locator to stop before exiting.  This method has problems with auto-reconnect and manual pull-the-plug testing showed that a locator would exit after a successful auto-reconnect.
> 1. start a locator with gfsh with enable-network-partition-detection
> 2. start a couple of servers on another machine
> 3. disconnect the locator machine's network cable
> 4. wait 30 seconds for the locator to detect the failure and go into auto-reconnect
> 5. reconnect the cable
> 6. observe the locator reconnect and exit



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