[
https://issues.apache.org/jira/browse/RATIS-176?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16412489#comment-16412489
]
Shashikant Banerjee commented on RATIS-176:
-------------------------------------------
Thanks [~szetszwo], for the review comments. I have a few doubts regarding the
comments.
1.The check should be in LogAppender.addEntry(..)
By moving the check, in LogAppender.addEntry in the server, what happens is,
createRequest fails , the sender is stopped and it triggers an re-election .
This goes on in an infinite loop. I think we should bail out instead of
retrying indefinitely.
When the clients sends a msg , the 1st log entry itself can be greater than the
maxBufferSize configured. In this case itself , the raft client itself can
detect and handle it.
> Log Appender should throw an Exception in case append entry size exceeds the
> maxBufferSize configured
> ------------------------------------------------------------------------------------------------------
>
> Key: RATIS-176
> URL: https://issues.apache.org/jira/browse/RATIS-176
> Project: Ratis
> Issue Type: Bug
> Reporter: Shashikant Banerjee
> Assignee: Shashikant Banerjee
> Priority: Minor
> Attachments: RATIS-176.001.patch
>
>
> LogAppender while adding append entry in LogEntryBuffer, checks whether the
> total allocated for all entries does not exceed the maxBufferSize allocated.
> In case, the size exceeds the limit ,entries are not added to the buffer but
> no exception is thrown . This case needs to be handled.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
