[
https://issues.apache.org/jira/browse/TINKERPOP-2359?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17086428#comment-17086428
]
Stephen Mallette commented on TINKERPOP-2359:
---------------------------------------------
I suppose that there is no reason that you could not call {{graph.close()}} by
sending a Gremlin script to the server to do so - that should work in
TinkerGraph's case. I guess you could also instead just make a {{g.io()}} call
and persist the data manually on the server. Of course, that would mean
changing how you initialize your graph when you restart - perhaps another
{{g.io()}} call in your server init script.
All that said, I'm curious as to why docker doesn't get you a nice clean
shutdown and of course how to make that work more nicely.
> onShutDown not being called when docker container stopped
> ---------------------------------------------------------
>
> Key: TINKERPOP-2359
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2359
> Project: TinkerPop
> Issue Type: Bug
> Components: server
> Affects Versions: 3.4.6
> Reporter: ohad israeli
> Priority: Minor
>
> When one uses groovy hooks with tinkerpop server hosted within the official
> docker the onStartUp hook will be called at server startup but the onShutDown
> will not be called.
> I am using tinkergraph server for local testing and i would like to persist
> the data upon the server exit. i've added a graph.close to the onShutDown
> hook but i havent been able to get it called.
> It will be called of course if you use the server start / server stop shell
> script but not if you kill / stop the docker container.
> If there is some other way to persist the date i.e. call the graph.close from
> code and note via the console it would also be great as using it from node
> javascript i havent been able to find a way to call the graph.close directly.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
