[infinispan-issues] [JBoss JIRA] (ISPN-5322) Support replicated cache for PH in a read-only mode

[Dan Berindei (JIRA)]

     [ https://issues.jboss.org/browse/ISPN-5322?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Dan Berindei updated ISPN-5322:
-------------------------------
              Status: Pull Request Sent  (was: Open)
    Git Pull Request: https://github.com/infinispan/infinispan/pull/3479


Indeed [~william.burns], partition handling will work with replicated caches.

Just like in distributed mode, there is no read-only mode, only degraded mode. 

A cache partition enters degraded mode if the current set of members does not contain a simple majority of the members in the latest stable cache topology, and the stable cache topology is updated when there is no pending rebalance. When a node joins, that means after the joiner receives the state, when a node leaves, it means after all the owners confirm the new topology.

> Support replicated cache for PH in a read-only mode
> ---------------------------------------------------
>
>                 Key: ISPN-5322
>                 URL: https://issues.jboss.org/browse/ISPN-5322
>             Project: Infinispan
>          Issue Type: Feature Request
>          Components: Core
>            Reporter: Wolf-Dieter Fink
>              Labels: partition_handling
>
> Current a REPL cache can not enable partition handling.
> There is a WARN message that the PH is ignored for replicated mode.
> The reason is that if one node leave the cache can become inconsistent.
> But for some reason it might be worth to have the cache available as read-only, so there is no way to make the cache inconsistent.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)