hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Flavio Paiva Junqueira (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-5188) Modifications to enable multiple types of logging
Date Sat, 30 May 2009 07:12:07 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-5188?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12714675#action_12714675
] 

Flavio Paiva Junqueira commented on HADOOP-5188:
------------------------------------------------

Konstantin, Could you be a little more specific on some of your comments? In particular, the
following is not clear to me:

# "But my main concern is that LogDevice (or LoggingDevice) is not the right abstraction.
And it seems to me that all necessary abstraction are already there you just need to use them."

You say that all necessary abstraction is there. But, where is "there"? What abstraction exactly
you believe the namenode has currently that enables one to keep the configuration and properties
of a device during the execution of the namenode? From what Luca is pointing out, it seems
that the abstractions you talk about are still implemented in a way that is very oriented
to files, and LogDevice enables one to abstract away how the log devices behave and are implemented.
Please shed some light on this one.

# "As I said before Edits Streams should be responsible for keeping information about logging
methods. And this is completely orthogonal to the number of BookKeeper clients you need to
instantiate."

I'm not sure what you're trying to say here. What does it mean to keep information about logging
methods? I believe they keep the methods used to access logs. Now, I agree that the number
of bookies that one needs to use is a separate issue, in the case of BookKeeper concretly.
However, we need to initialize ledgers, and for that we need to remember configuration parameters
for BookKeeper and we need a different initialization procedure. Where should we keep config
parameters for BookKeeper and where should we instantiate and keep a BookKeeper object?

Also, I agree with your comment on the use of "instanceof".

 

> Modifications to enable multiple types of logging 
> --------------------------------------------------
>
>                 Key: HADOOP-5188
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5188
>             Project: Hadoop Core
>          Issue Type: Improvement
>          Components: dfs
>    Affects Versions: 0.20.0
>            Reporter: Luca Telloli
>             Fix For: 0.21.0
>
>         Attachments: HADOOP-5188.patch, HADOOP-5188.patch, HADOOP-5188.patch, HADOOP-5188.patch,
HADOOP-5188.patch, HADOOP-5188.patch, HADOOP-5188.pdf
>
>


-- 
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