[
https://issues.apache.org/jira/browse/DERBY-5565?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Kathey Marsden updated DERBY-5565:
----------------------------------
Attachment: derby-5565_diff.txt
Attached is a patch for this issue.
We now reject all PRDID's that do not start with DNC.
The protocol tests have been changed to use the DNC10090 PRDID instead of
TST01000. This actually alters the code path where there is slightly different
behavior in later derby client versions, so that we have to skip an additional
dss on connect for the diagnostic information. After this commit I plan to add
another protocol test for the client reject but thougth I would get just the
changes for the existing tests in first.
Tests passed. Please review.
> Network Server should reject client connections that are not Derby Network
> Client
> ---------------------------------------------------------------------------------
>
> Key: DERBY-5565
> URL: https://issues.apache.org/jira/browse/DERBY-5565
> Project: Derby
> Issue Type: Improvement
> Components: Network Server
> Affects Versions: 10.9.0.0
> Reporter: Kathey Marsden
> Assignee: Kathey Marsden
> Attachments: JCCDSConnectTest.java, derby-5565_diff.txt,
> derby-5565_diff.txt, derby.log
>
>
> Since there have been no other network clients besides Derby Network Client
> tested or supported with Derby since 10.1 and since any protocol based client
> needs to understand Derby's DRDA extensions, deviations, and stored procedure
> usage. I think it would be a good idea in 10.9 for Network Server to
> outright reject any network clients that are not Derby Network Client.
> This would eliminate confusion up front for those that might not be aware
> that the DB2 Universal JDBC Driver and DB2 Runtime Client are not supported.
> They would get a clean reasonable error instead of hitting various protocol
> errors.
> Also it would mean if someone does want to add support for some network
> client in the future they would at least need to add the one or two lines of
> code in AppRequester to identify it, which I think would be a good thing.
> I think the code change would not be hard but the biggest impact might be
> anyone who still runs tests with JCC on trunk would need to disable those
> tests. There is a separate issue DERBY-4785 that Jayaram is working on to
> complete remove the JCC related code from the tests and test infrastructure.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
