[
https://issues.apache.org/jira/browse/HBASE-15264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15149863#comment-15149863
]
stack commented on HBASE-15264:
-------------------------------
Sweet. Stick on head of the class too.
How we know when to flush? We'll do it size and/or time based? We'll plug this
in on what was the append/sync side of the ringbuffer?
On error, we need to block and wait on the lease recovery? Or is there where we
do as you suggest, we put the WAL aside while recovering lease and start a new
one. What you thinking for case where we fail to recover lease? Or, you are
thinking we accumulate writes in memory without acking the client while lease
recovery is going on (Pardon me if missing some of what has been discussed
before... I tried to catch up today but these issues are getting long now --
smile).
Great stuff [~Apache9]
> Implement a fan out HDFS OutputStream
> -------------------------------------
>
> Key: HBASE-15264
> URL: https://issues.apache.org/jira/browse/HBASE-15264
> Project: HBase
> Issue Type: Sub-task
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Attachments: HBASE-15264-v1.patch, HBASE-15264-v2.patch,
> HBASE-15264.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
