[
https://issues.apache.org/jira/browse/HDFS-13873?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16719539#comment-16719539
]
Erik Krogen commented on HDFS-13873:
------------------------------------
The logic seems fine to me now that the exception handling has been fixed. I
agree with your previous comments regarding attackers.
* A few of the "long" lines you shortened were actually exactly 80 characters,
i.e. not too long: {{ClientGSIContext}} L64 and {{GlobalStateIdContext}} L56
and L67
* The new constants in {{AlignmentContext}} could probably use a Javadoc?
* {{GlobalStateIdContext}} constructor has an extra parameter listed in its
Javadoc
* {{GlobalStateIdContext#receieveRequestState()}}, in the existing
if-condition, you modified it to return {{serverStateId}} whereas previously it
would return {{clientStateId}}. This seems to violate the Javadoc which states
that the return is "state id of in the request header."
* The name, and Javadoc description, for the new parameter on
{{#receiveRequestState()}} do not match between {{GlobalStateIdContext}} and
{{AlignmentContext}}. It seems like the one on {{GlobalStateIdContext}} is the
correct one, but we should also update it to include units in the name and/or
Javadoc
* {{(clientWaitTime / 1000)}} -> can we use
{{TimeUnit.MILLISECONDS.toSeconds(clientWaitTime)}}?
* In {{HATestUtil}}, it seems like adding a new package-private method
{{setLastSeenStateIdForTest()}} would be more appropriate than the reflection
used here, thoughts?
> ObserverNode should reject read requests when it is too far behind.
> -------------------------------------------------------------------
>
> Key: HDFS-13873
> URL: https://issues.apache.org/jira/browse/HDFS-13873
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs-client, namenode
> Affects Versions: HDFS-12943
> Reporter: Konstantin Shvachko
> Assignee: Konstantin Shvachko
> Priority: Major
> Attachments: HDFS-13873-HDFS-12943.001.patch,
> HDFS-13873-HDFS-12943.002.patch, HDFS-13873-HDFS-12943.003.patch
>
>
> Add a server-side threshold for ObserverNode to reject read requests when it
> is too far behind.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
