[
https://issues.apache.org/jira/browse/JENA-214?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13214533#comment-13214533
]
Paolo Castagna commented on JENA-214:
-------------------------------------
> Not for readWriteGraphStoreEP as it is both read and write operations.
Yep. Good point.
> Can holders be restarted?
There is an holder.start() method as well, so I suppose so.
I have never used/done it before, so I am not 100% sure it is a safe/good thing
to do.
> Won't programming based on the lifecycle will make it Jetty specific? - see
> also JENA-201 which is asking for Fuseki as a (portable) WAR file.
Right, good point... back to thinking mode... I'll look if there a generic way
given a path /{dataset}/{service} to find the servlet instance dealing with
those requests. If that is possible, we could have a flag to switch between
read-write to read-only mode on the necessary servlets.
> Allow to set Fuseki (or a single dataset) in read-only mode (and eventually,
> back into read-write mode)
> -------------------------------------------------------------------------------------------------------
>
> Key: JENA-214
> URL: https://issues.apache.org/jira/browse/JENA-214
> Project: Apache Jena
> Issue Type: New Feature
> Components: Fuseki
> Reporter: Paolo Castagna
> Priority: Minor
> Attachments: ActionReadOnly.java
>
>
> A very useful MgtCmdServlet with commands such as: "backup", "shutdown", etc.
> is now available in Fuseki.
> This is proposing a new command to set the server (or a single dataset) in
> read-only mode and back into read-write mode.
> A use case for this would be to perform a replication via rsync, in order to
> replicate TDB indexes from server A to server B both servers needs to be set
> to read-only mode first, the replication can start, when it finishes both
> servers can be configured back to read-write mode.
> Another use case would be, enabling a backup with an external tool or any
> other operation which needs to have exclusive access to TDB indexes and/or
> ensure all data has been flushed to disk and there are no in-flight write
> transactions.
> See also a relevant message from jena-dev mailing list:
> -
> http://mail-archives.apache.org/mod_mbox/incubator-jena-dev/201201.mbox/%3C4F1F1C9B.9050309%40apache.org%3E
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
