hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Raghu Angadi (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-1134) Block level CRCs in HDFS
Date Tue, 03 Apr 2007 20:42:32 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-1134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12486492
] 

Raghu Angadi commented on HADOOP-1134:
--------------------------------------


Not generating missing CRCs sounds fine and in the initial release datanode will not delete
any blocks for checksum reasons.

> That assumes that the checksum is validated, doesn't it? And we're not yet validating
checksums except in the client,

Please confirm the desired behaviour in normal operation (after upgrade) : Two options I have
seen till now:

1) Datanode does not verify any checksums anytime. This might or might not change in future.

2) Datanode also verifies checksums while receiving and sending data. It closes the client
connection if checksum does not match while receiving from client. In short term, it at most
prints a warning for checksum errors while sending data and does not report the error.

Doug, I guess you are proposing (1). Though I prefer (2), I am ok with (1). If this is not
true I would like to know what the desired behavior is.



> 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.


Mime
View raw message