[
https://issues.apache.org/jira/browse/HBASE-7533?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13556636#comment-13556636
]
Enis Soztutar commented on HBASE-7533:
--------------------------------------
bq. First, let me say, using exceptions for messaging is perverse as
RegionMovedException is doing.
That is why I am more convinced of the alternate proposal of having things like
RME as valid return types from get/put/etc.
bq. Second, how many instances of payload carrying exceptions do we have? One,
two?
The most offending one I can think of is DoNotRetryIOException. By definition,
you are passing some semantics (although no data) to the client as an exception
by using the exception class hierarchy. Imagine again a C client, in which we
get the actual exception class, but don't have the java class hierarchy to
reason about whether this class inherits from DNRIOE.
> Write an RPC Specification for 0.96
> -----------------------------------
>
> Key: HBASE-7533
> URL: https://issues.apache.org/jira/browse/HBASE-7533
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Assignee: stack
> Fix For: 0.96.0
>
> Attachments: 7533_proto_sketch.txt
>
>
> RPC format is changing for 0.96 to accomodate our protobufing all around.
> Here is a first cut. Please shred:
> https://docs.google.com/document/d/1-1RJMLXzYldmHgKP7M7ynK6euRpucD03fZ603DlZfGI/edit
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
