[ 
https://issues.apache.org/jira/browse/DERBY-3417?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Dag H. Wanvik updated DERBY-3417:
---------------------------------

    Attachment: derby-3417.stat
                derby-3417.diff

Uploading a trial patch that makes the seen SQL states session level severity.
ReplicationSuite still works, some of the time... :( That is, I do see 
intermittent errors,
but I *think* none of them are new with this patch.

I replaced imports of SQLState; tests should be self contained.
I also replaced some assertTrue with assertSQLState for better error reporting.

I am still not clear on if any of these errors could be thrown in contexts 
where it would be wrong to give them session level severity; Knut's proposal 
would avoid that issue, at the cost of a new mechanism (already enough in the 
error apparatus, perhaps :)

I did not make all the XRE* errors session level yet, maybe they could all be 
given that severity?


> slave side stop in a client server mode results in SQLState printed without 
> proper error message
> ------------------------------------------------------------------------------------------------
>
>                 Key: DERBY-3417
>                 URL: https://issues.apache.org/jira/browse/DERBY-3417
>             Project: Derby
>          Issue Type: Bug
>          Components: Replication
>    Affects Versions: 10.4.1.3
>            Reporter: V.Narayanan
>            Assignee: Dag H. Wanvik
>         Attachments: derby-3417.diff, derby-3417.stat
>
>
> I tried a stopSlave on the slave side of the replication system and
> found the below
> ij> connect 'jdbc:derby://localhost:1528/replicationdb;stopSlave=true';
> ERROR XRE41: DERBY SQL error: SQLCODE: -1, SQLSTATE: XRE41, SQLERRMC: XRE41
> https://issues.apache.org/jira/browse/DERBY-3205 says
> ERROR XRE41: Replication operation 'failover' or 'stopSlave' failed because 
> the connection with the master is working. Issue the 'failover' or 
> 'stopMaster' operation on the master database instead.
> needs  to be printed.
> I am not sure if this is a generic case for client server replication 
> messages.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to