[
https://issues.apache.org/jira/browse/HDFS-11644?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16110224#comment-16110224
]
Sean Busbey commented on HDFS-11644:
------------------------------------
We need docs on using this feature. I'd like to file a blocker on HBase 2.0 to
make use of it where it is present, specifically so that we can avoid deploying
the HBase WAL on filesystems that don't support hflush/hsync (such as EC). The
lack of ability to reliably flush/sync doesn't appear to be called out in the
Erasure Coding docs as a concern for those who might deploy it and I'd rather
HBase bring the risk to operators' attention by failing fast when it finds the
configuration.
But I can't find any guidance on using this feature (or that it exists) in the
3.0.0-alpha4 release notes nor in the filesystem docs. I also can't tell from
the javadocs on StreamCapabilities what I'm even supposed to query for to check
on the fundamentals that this jira is addressing (hflush / hsync).
> Support for querying outputstream capabilities
> ----------------------------------------------
>
> Key: HDFS-11644
> URL: https://issues.apache.org/jira/browse/HDFS-11644
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
> Affects Versions: 3.0.0-alpha1
> Reporter: Andrew Wang
> Assignee: Manoj Govindassamy
> Labels: hdfs-ec-3.0-must-do
> Fix For: 2.9.0, 3.0.0-alpha4
>
> Attachments: HDFS-11644.01.patch, HDFS-11644.02.patch,
> HDFS-11644.03.patch, HDFS-11644-branch-2.01.patch
>
>
> FSDataOutputStream#hsync checks if a stream implements Syncable, and if so,
> calls hsync. Otherwise, it just calls flush. This is used, for instance, by
> YARN's FileSystemTimelineWriter.
> DFSStripedOutputStream extends DFSOutputStream, which implements Syncable.
> However, DFSStripedOS throws a runtime exception when the Syncable methods
> are called.
> We should refactor the inheritance structure so DFSStripedOS does not
> implement Syncable.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
