[ https://issues.apache.org/jira/browse/HDFS-744?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13016626#comment-13016626
]
Hairong Kuang commented on HDFS-744:
------------------------------------
I thought more about this. It seems to me it is not easy to support hsync. If the bytes written
in between two hsync calls are across a block boundary. The first block needs to get fsynced
to disk as well.
An alternative proposal is to provide a filesystem create API that takes a fsync option. If
fsync is true, any hflush call will force the data to be synced to disk and any block will
be synced to disk upon close as well.
> Support hsync in HDFS
> ---------------------
>
> Key: HDFS-744
> URL: https://issues.apache.org/jira/browse/HDFS-744
> Project: Hadoop HDFS
> Issue Type: New Feature
> Reporter: Hairong Kuang
>
> HDFS-731 implements hsync by default as hflush. As descriibed in HADOOP-6313, the real
expected semantics should be "flushes out to all replicas and all replicas have done posix
fsync equivalent - ie the OS has flushed it to the disk device (but the disk may have it in
its cache)." This jira aims to implement the expected behaviour.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
|