[
https://issues.apache.org/jira/browse/HADOOP-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12486444
]
Raghu Angadi commented on HADOOP-1134:
--------------------------------------
> If the only copy of a block has no CRC, shouldn't we still permit folks to
> access the data somehow? I don't think we should just remove the block in
> that case.
Agreeed. earlier in the Jira I mentioned that Namnode should be enhanced to
treat 'block corrupt' message from Datanode properly. i.e. try to create a new
replica and not delete the block until a new replica can be created.
I wanted to imply if CRC file does not exist, it will be treated just like a
corrupt block (whatever the policy may be). I somehow thought you meant to
distinguish between 'missing checksum' and 'checksum mismatch'.
> access the data even if it doesn't have a checksum. Wouldn't that be
> preferable to data loss?
yes. This will be built into client-datanode data transfer protocol.
> Block level CRCs in HDFS
> ------------------------
>
> Key: HADOOP-1134
> URL: https://issues.apache.org/jira/browse/HADOOP-1134
> Project: Hadoop
> Issue Type: New Feature
> Components: dfs
> Reporter: Raghu Angadi
> Assigned To: Raghu Angadi
>
> Currently CRCs are handled at FileSystem level and are transparent to core
> HDFS. See recent improvement HADOOP-928 ( that can add checksums to a given
> filesystem ) regd more about it. Though this served us well there a few
> disadvantages :
> 1) This doubles namespace in HDFS ( or other filesystem implementations ). In
> many cases, it nearly doubles the number of blocks. Taking namenode out of
> CRCs would nearly double namespace performance both in terms of CPU and
> memory.
> 2) Since CRCs are transparent to HDFS, it can not actively detect corrupted
> blocks. With block level CRCs, Datanode can periodically verify the checksums
> and report corruptions to namnode such that name replicas can be created.
> We propose to have CRCs maintained for all HDFS data in much the same way as
> in GFS. I will update the jira with detailed requirements and design. This
> will include same guarantees provided by current implementation and will
> include a upgrade of current data.
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
