[
https://issues.apache.org/jira/browse/HDFS-12911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16328428#comment-16328428
]
Rakesh R commented on HDFS-12911:
---------------------------------
Thanks [~umamaheswararao] for the patch. Adding few comments.
# Please update the javadoc of \{{StoragePolicySatisfier}} class, now it is
representing only internal. Good to make it generic for both the cases.
{code}
Here Namenode
* will pick the file blocks which are expecting to change its storages
{code}
# Make spsPaths as final - \{{private final SPSPathIds spsPaths;}}. Also, can
you keep this attribute after \{{private boolean spsEnabled;}}, this will help
in rebasing. Thanks!
# How abt keeping the block move logic inside
IntraSPSNameNodeBlockMoveTaskHandler class instead of BlockManager.java. Just a
thought to keep minimal changes in BlockManager.
{code}
/**
* Assigns the block movement task to target datanode.
*/
@Override
public void submitMoveTask(BlockMovingInfo blkMovingInfo,
BlockMovementListener blockMoveCompletionListener) throws IOException {
namesystem.readLock();
try {
DatanodeDescriptor dn = datanodeManager
.getDatanode(blkMovingInfo.getTarget().getDatanodeUuid());
if (dn == null) {
throw new IOException("Failed to schedule block movement task:"
+ blkMovingInfo + " as target datanode: "
+ blkMovingInfo.getTarget() + " doesn't exists");
}
dn.addBlocksToMoveStorage(blkMovingInfo);
dn.incrementBlocksScheduled(blkMovingInfo.getTargetStorageType());
} finally {
namesystem.readUnlock();
}
}
{code}
# Add \{{@InterfaceAudience.Private}} to BlockMovementListener, SPSPathIds,
ItemInfo.
# Typo \{{pathIDProcesor}} => \{{pathIDProcessor}}
> [SPS]: Modularize the SPS code and expose necessary interfaces for
> external/internal implementations.
> -----------------------------------------------------------------------------------------------------
>
> Key: HDFS-12911
> URL: https://issues.apache.org/jira/browse/HDFS-12911
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Reporter: Uma Maheswara Rao G
> Assignee: Uma Maheswara Rao G
> Priority: Major
> Attachments: HDFS-12911-HDFS-10285-01.patch,
> HDFS-12911-HDFS-10285-02.patch, HDFS-12911-HDFS-10285-03.patch,
> HDFS-12911.00.patch
>
>
> One of the key comment from the discussions was to modularize the SPS code,
> so we can easily plug in the external/internal implementations. This JIRA for
> doing the necessary refactoring.
> So other comments to handle
> Daryn:
> # Lock should not kept while executing placement policy.
> - handled by HDFS-12982
> # While starting up the NN, SPS Xattrs checks happen even if feature
> disabled. This could potentially impact the startup speed.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
