[
https://issues.apache.org/jira/browse/CASSANDRA-5290?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marcus Eriksson updated CASSANDRA-5290:
---------------------------------------
Summary: Refactor inter-node SSL support to make it possible enable without
total cluster downtime (was: Refactor SSL support to make it possible enable
without total cluster downtime)
> Refactor inter-node SSL support to make it possible enable without total
> cluster downtime
> -----------------------------------------------------------------------------------------
>
> Key: CASSANDRA-5290
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5290
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Marcus Eriksson
> Priority: Minor
>
> should be possible by doing something similar to what compression does today:
> * node A connects to node B, indicating that it wants SSL
> * node A and B upgrade/wrap sockets
> * node B must still be able to talk to A (and all other nodes) without SSL
> nothing secret is shared during hand shake phase
> one difference compared to compression is that after the cluster is rolled
> there must be a way to not allow any non-ssl talk at all to avoid any
> plaintext-communication due to configuration mistakes. this should be doable
> via configuration and perhaps JMX
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira