[jira] [Comment Edited] (CASSANDRA-8377) Coordinated Commitlog Replay

["T Jake Luciani (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-8377?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14610990#comment-14610990 ] 

T Jake Luciani edited comment on CASSANDRA-8377 at 7/1/15 8:56 PM:
-------------------------------------------------------------------

Quick initial review

 * Can you write a dtest for this?  perhaps move the CL out of the way and replay them from the wrong nodes.
 * There seems to be a lot of code that assumes system properties exist? can you assert this?
 * I think when you replay you shouldn't mutate at CL.ONE but mutate with the batchlog at quorum. so even if the replay fails it will be in the batchlog so you can continue the replay (and replay should be made so it wont fail on a mutate() exception. 
 * This should be accessible from nodetool


was (Author: tjake):
Quick initial review

 * Can you write a dtest for this?  perhaps move the CL out of the way and replay them from the wrong nodes.
 * There seems to be a lot of code that assumes system properties exist? can you assert this?
 * I think when you replay you shouldn't mutate at CL.ONE but mutate with the batchlog at quorum. so even if the replay fails it will be in the batchlog so you can continue the replay (and replay should be made so it wont fail on a mutate() exception. 

> Coordinated Commitlog Replay
> ----------------------------
>
>                 Key: CASSANDRA-8377
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8377
>             Project: Cassandra
>          Issue Type: New Feature
>            Reporter: Nick Bailey
>            Assignee: Chris Lohfink
>             Fix For: 3.x
>
>         Attachments: CASSANDRA-8377.txt
>
>
> Commit log archiving and replay can be used to support point in time restores on a cluster. Unfortunately, at the moment that is only true when the topology of the cluster is exactly the same as when the commitlogs were archived. This is because commitlogs need to be replayed on a node that is a replica for those writes.
> To support replaying commitlogs when the topology has changed we should have a tool that replays the writes in a commitlog as if they were writes from a client and will get coordinated to the correct replicas.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)