geode-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Fred Krone (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (GEODE-3294) Spike: investigate and recommend best way to mitigate stale data
Date Tue, 25 Jul 2017 15:28:00 GMT

     [ https://issues.apache.org/jira/browse/GEODE-3294?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Fred Krone updated GEODE-3294:
------------------------------
    Description: 
There are many ways to invalidate stale data in a cache.  

Should the cluster hold DB credentials to call a delta table of updated keys to invalidate?
Should the client initiate cache restored state?  etc
What about a split brain scenario where the network goes down but GemFire is not recovering?
 
Should it block reads during recovery but allow writes immediately?

  was:
There are many ways to invalidate stale data in a cache.  

Should the cluster hold DB credentials to call a delta table of updated keys to invalidate?
Should the client initiate cache restored state?  etc
What about a split brain scenario where the network goes down but GemFire is not recovering?
 


> Spike: investigate and recommend best way to mitigate stale data
> ----------------------------------------------------------------
>
>                 Key: GEODE-3294
>                 URL: https://issues.apache.org/jira/browse/GEODE-3294
>             Project: Geode
>          Issue Type: Sub-task
>          Components: persistence
>            Reporter: Fred Krone
>
> There are many ways to invalidate stale data in a cache.  
> Should the cluster hold DB credentials to call a delta table of updated keys to invalidate?
> Should the client initiate cache restored state?  etc
> What about a split brain scenario where the network goes down but GemFire is not recovering?
 
> Should it block reads during recovery but allow writes immediately?



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message