[
https://issues.apache.org/jira/browse/HADOOP-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12485345
]
Doug Cutting commented on HADOOP-1134:
--------------------------------------
> The client would need a way to go from a block id to a .crc file via an
> extension of the Namenode API.
The split could include the datanode name, the block ID, the file name and the
offset of the block within the file. Then the mapper could access the CRC file
using normal namenode and datanode calls.
But actually, now that I think about it, if we're primarily not validating
checksums against the data, but rather comparing all the checksums for a block,
then locality may not be worthwhile. In that case we'd want a temporary
datanode extension that permits writing the checksum file for a block. Then
the updater map task can read through all copies of the checksum file,
construct the best possible checksum for each block, then send these to
datanodes. So, in aggregate, 6% of the filesystem would cross the wire during
the upgrade. Could that work?
> Once a file is validated the .crc file would be deleted by the client.
Is the upgrade the time to detect corrupt blocks? Won't these be detected
through the normal mechanisms later? We don't want to perform any replication
during the upgrade. As a subsequent patch, we should validate checksums during
replication, so that we don't replicate a corrupt block, but I don't think we
need to do that as a part of this patch. (Some might wait until both patches
are committed before updating a particular filesystem.)
> 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.
