[
https://issues.apache.org/jira/browse/HBASE-15537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15225727#comment-15225727
]
Duo Zhang commented on HBASE-15537:
-----------------------------------
[~stack] I add the following debug log in {{AbstractFSWAL.rollWriter}}.
{code:title=AbstractFSWAL.java}
if (LOG.isDebugEnabled()) {
LOG.debug("Create new " + getClass().getSimpleName() + " writer with
pipeline: "
+ Arrays.toString(getPipeline()));
}
{code}
Is it enough to get the statistics you want?
HDFS does have a excludeNodes and a favoredNodes when allocating new blocks,
but we need a good abstraction first because the multiwal is not bound with
HDFS I think? It should work with any WAL implementations. And also, we could
not control other regionservers on the DN selection so I wonder whether it can
perform better than the default selection algorithm on a real large cluster.
Thanks.
> Add multi WAL support for AsyncFSWAL
> ------------------------------------
>
> Key: HBASE-15537
> URL: https://issues.apache.org/jira/browse/HBASE-15537
> Project: HBase
> Issue Type: Sub-task
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-15537-v3.patch, HBASE-15537-v4.patch,
> HBASE-15537-v5.patch, HBASE-15537.patch, HBASE-15537_v2.patch
>
>
> The multi WAL should not be bound with {{FSHLog}}.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
