[
https://issues.apache.org/jira/browse/HDFS-13658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16560404#comment-16560404
]
Xiao Chen commented on HDFS-13658:
----------------------------------
Thanks for revving Kitti.
- ECBlockGroupStats and ReplicatedBlockStats: I think it's semantically better
to make the new field final. (addXXX isn't a good name when it's doing a
'set'). I think it's ok here to just add an overload to the ctor, given builder
a more tedious change and we don't expect the stats to change often in the
future.
- Sorry I didn't bring this up in the last comment. See
http://hadoop.apache.org/docs/r3.0.0/hadoop-project-dist/hadoop-common/Compatibility.html
for a full compat guideline. The new field in {{ClientNamenodeProtocol.proto}}
need to be optional to be wire-compatible with older versions.
- With the proto change, in {{PBHelperClient}} you can check the proto
{{hasXXX}}, and if so, {{getXXX}}.
- minor, we can rename {{isHighestPriorityBlocksStatSet}} to
{{hasHighestPriorityLowRedundancyBlocks}} to match protobuf.
- Could you clarify about
{{TestLowRedundancyBlockQueues#doTestStripedBlockPriorities}} a little? I was
hoping to see the test where [1, parityBlkNum) of corrupt blocks doesn't change
highest priority queue, and the last (parityBlkNum) corruption would cause the
queue to increase - but I may be missing something
> fsck, dfsadmin -report, and NN WebUI should report number of blocks that have
> 1 replica
> ---------------------------------------------------------------------------------------
>
> Key: HDFS-13658
> URL: https://issues.apache.org/jira/browse/HDFS-13658
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs
> Affects Versions: 3.1.0
> Reporter: Kitti Nanasi
> Assignee: Kitti Nanasi
> Priority: Major
> Attachments: HDFS-13658.001.patch, HDFS-13658.002.patch,
> HDFS-13658.003.patch, HDFS-13658.004.patch, HDFS-13658.005.patch,
> HDFS-13658.006.patch, HDFS-13658.007.patch, HDFS-13658.008.patch,
> HDFS-13658.009.patch
>
>
> fsck, dfsadmin -report, and NN WebUI should report number of blocks that have
> 1 replica. We have had many cases opened in which a customer has lost a disk
> or a DN losing files/blocks due to the fact that they had blocks with only 1
> replica. We need to make the customer better aware of this situation and that
> they should take action.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
