[
https://issues.apache.org/jira/browse/HDFS-611?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Shao updated HDFS-611:
----------------------------
Attachment: HDFS-611.trunk.v5.patch
bq. It would look really good if you could move the deleteAsync() method
together with the static class ReplicaFileDeleteClass inside AsyncDiskService.
I understand your motivation that you want to keep AsyncDiskService independent
of data-node connotations if the goal is to reuse it in MR, but this could be
done by simple factoring our the common base class if such usage in MR will
ever materialize.
For the MR usage, I am think about moving the AsyncDiskService class directly
to common. Aggregation seems better than inheritance here, just like ThreadPool
(I guess we don't extend ThreadPoolExecutor for different types of Tasks).
It seems to me that AsyncDiskService should not need to have knowledge of
whatever Task is requested by the caller. In particular, the decDfsUsage call
in the ReplicaFileDeletionTask is closely related to FSDataSet, and should be
maintain inside FSDataSet. What do you think?
bq. A nit: could you convert the comment for AsyncDiskService to JavaDoc
comment by just adding one more star.
Added the missing "*" for javadoc.
> Heartbeats times from Datanodes increase when there are plenty of blocks to
> delete
> ----------------------------------------------------------------------------------
>
> Key: HDFS-611
> URL: https://issues.apache.org/jira/browse/HDFS-611
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: data-node
> Affects Versions: 0.20.1, 0.21.0, 0.22.0
> Reporter: dhruba borthakur
> Assignee: Zheng Shao
> Fix For: 0.20.2, 0.21.0, 0.22.0
>
> Attachments: HDFS-611.branch-19.patch, HDFS-611.branch-19.v2.patch,
> HDFS-611.branch-20.patch, HDFS-611.branch-20.v2.patch, HDFS-611.trunk.patch,
> HDFS-611.trunk.v2.patch, HDFS-611.trunk.v3.patch, HDFS-611.trunk.v4.patch,
> HDFS-611.trunk.v5.patch
>
>
> I am seeing that when we delete a large directory that has plenty of blocks,
> the heartbeat times from datanodes increase significantly from the normal
> value of 3 seconds to as large as 50 seconds or so. The heartbeat thread in
> the Datanode deletes a bunch of blocks sequentially, this causes the
> heartbeat times to increase.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
