[
https://issues.apache.org/jira/browse/HBASE-7533?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13556843#comment-13556843
]
Enis Soztutar commented on HBASE-7533:
--------------------------------------
bq. But I'd suggest we do not need a generically extensible system that allows
adding new payload-bearing exception types (see above where we would want to
discourage control via exception).
Totally agreed.
bq. Do you lot think there could be more? It would have to be exceptions that
were not method specific because for these types then we should suggest the
method response include the exception as an optional response. So that leaves
exceptions that could come out of multiple methods or problems the server
encounters while running.
Again, the other concerning thing is DoNotRetryIOException. Although it does
not carry data, we basically do control flow on subclasses of this exception.
Maybe we should add a field bool doNotRetry, in the RpcException.
bq. I'd think that we'd try and keep the spec as narrow as possible (and no
narrower) so if we could do without having an extensible fancy payload bearing
exception types system, lets punt.
Agreed. But what I am saying is that we already do have those in place and get
rid of the string parsing one way or the other.
> 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_spec.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
