[
https://issues.apache.org/jira/browse/DERBY-3064?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12530410
]
V.Narayanan commented on DERBY-3064:
------------------------------------
Thank you for the comments and the reviews oystein.
>4. Why are exceptions from shipALogChunk handled differently by run()
> and forceFlush()? Maybe you could let shipALogChunk handle the
> exceptions instead so you do not have to do it in two places?
* It is not possible to throw exceptions from the run method, hence I throw it
as
RuntimeExceptions.
* forceFlush is a method that might have a different implementation if in future
we decide we are going to implement a more sophisticated LogShipper, but the
basic interface should remain common. So the exception that is thrown should
be the same irrespective of the implementation.
So the best way of making the exceptions thrown in derby uniform is using the
Derby StandardException
* I thought about handling the exceptions in shipALogChunk itself, but we need
some way of informing the MasterController that the shipping has failed. We
need
the MasterController to be informed of the exception.
Handling the exceptions inside shipALogChunk would muffle the exceptions
inside
this method.
The above three considerations were what forced me to go in for the
heterogeneity
in handling the exceptions between the two methods.
> Implement the LogShipper that will enable the shipping of Log records from
> the master to the slave
> --------------------------------------------------------------------------------------------------
>
> Key: DERBY-3064
> URL: https://issues.apache.org/jira/browse/DERBY-3064
> Project: Derby
> Issue Type: Sub-task
> Reporter: V.Narayanan
> Assignee: V.Narayanan
> Attachments: LogShipperImpl_v1.diff, LogShipperImpl_v1.stat,
> LogShipperImpl_v2.diff, LogShipperImpl_v2.stat
>
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
