You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@geode.apache.org by "Jared Stewart (JIRA)" <ji...@apache.org> on 2016/09/29 17:41:20 UTC

[jira] [Updated] (GEODE-1950) Start locator hangs forever on AWS when specifying -Djava.rmi.server.hostname

     [ https://issues.apache.org/jira/browse/GEODE-1950?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jared Stewart updated GEODE-1950:
---------------------------------
    Attachment: locatorThreadDump.txt
                gfshThreadDump.txt

I have attached the thread dumps for both the gfsh and the locator processes.  It looks like gfsh is attempting to make a JMX connection to the locator via its external IP address.  Since the default AWS configuration does not allow inbound connections to the JMX port, gfsh is of course unable to establish this connection.  However, the connection attempt should time out rather than hanging indefinitely.

> Start locator hangs forever on AWS when specifying -Djava.rmi.server.hostname 
> ------------------------------------------------------------------------------
>
>                 Key: GEODE-1950
>                 URL: https://issues.apache.org/jira/browse/GEODE-1950
>             Project: Geode
>          Issue Type: Bug
>          Components: management
>            Reporter: Jared Stewart
>         Attachments: gfshThreadDump.txt, locatorThreadDump.txt
>
>
> If you spin up an AWS instance, specify {{--J=-Djava.rmi.server.hostname=<public_ip>}}, and don't open incoming tcp connections for the JMX port in the AWS security group for that instance, the start locator command hangs indefinitely.  
> {code}
> gfsh>start locator --name=l1 --J=-Djava.rmi.server.hostname=<public_ip>
> {code}



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