[jira] [Commented] (RATIS-1262) Support linearizable read

["Glen Geng (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/RATIS-1262?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17254445#comment-17254445 ] 

Glen Geng commented on RATIS-1262:
----------------------------------

I suggested to support leader with lease by enhance RATIS-981, so that we can practically avoid the split-brain issue in the raft cluster (leaders of two different terms can not exist together).

> Support linearizable read
> -------------------------
>
>                 Key: RATIS-1262
>                 URL: https://issues.apache.org/jira/browse/RATIS-1262
>             Project: Ratis
>          Issue Type: New Feature
>            Reporter: runzhiwang
>            Assignee: runzhiwang
>            Priority: Major
>         Attachments: 截屏2020-12-24 下午3.31.32.png
>
>
> There are two case violate linearizable read, for example, there are 3 servers: server1, server2, server3, and server1 is leader
> 1.  server1: applyIndex = 2, commitIndex = 2, server2: applyIndex = 1, commitIndex = 2. When server1 step down, server2 become leader, and client read from server2,
> because server2's applyIndex less than commitIndex, client will read old data.
> 2.  when split-brain happens, server2 was elected as new leader, but server1 still think it's leader, when client read from server1, if server2 has processed write request, client will read old data from server1. As [~glengeng] explained, RATIS-981 still exist dirty-read in the min election timeout if split-brain happens.
> etcd / sofa-jraft / pingcap all support linearizable read which has been explained in raft paper, we can implement it in ratis.  It's important in Ozone HA.
> pingcap's linearizable read: https://en.pingcap.com/blog/lease-read



--
This message was sent by Atlassian Jira
(v8.3.4#803005)