| From: | Kris Jurka <books(at)ejurka(dot)com> |
|---|---|
| To: | Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov> |
| Cc: | Donald Fraser <postgres(at)kiwi-fraser(dot)net>, "[JDBC]" <pgsql-jdbc(at)postgresql(dot)org> |
| Subject: | Re: Invalid PSQLState used in connection classes |
| Date: | 2011-03-19 23:47:31 |
| Message-ID: | alpine.BSO.2.00.1103191936050.28447@leary.csoft.net |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-jdbc |
On Tue, 8 Mar 2011, Kevin Grittner wrote:
> "Donald Fraser" <postgres(at)kiwi-fraser(dot)net> wrote:
>
>> However if I have a broken network between my client and the
>> server, such that there is no route to the server, I can get a
>> ConnectException and I have not established, in any way, a
>> connection to the PostgreSQL server.
>
> 08001 - SQLCLIENT UNABLE TO ESTABLISH SQLCONNECTION
>
> Sounds better for that sort of failure.
>
Sounds reasonable to me. Looking at the error codes used during startup,
I think we should also make some other changes at the same time.
CONNECTION_UNABLE_TO_CONNECT -> PROTOCOL_VIOLATION
CONNECTION_FAILURE -> PROTOCOL_VIOLATION
when the driver receives an unknown message.
CONNECTION_FAILURE -> CONNECTION_REJECTED
when the server doesn't support SSL, but we need a SSL connection. I
think CONNECTION_FAILURE should only be used when something has gone wrong
after a connection has been established.
Patch attached.
Kris Jurka
| Attachment | Content-Type | Size |
|---|---|---|
| connection-sqlstates.patch | text/plain | 12.7 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kris Jurka | 2011-03-20 00:12:15 | Re: Failing test in org.postgresql.test.jdbc2.DateTest |
| Previous Message | Kris Jurka | 2011-03-19 22:49:56 | Re: unwrapping InvocationTargetException in PGXAConnection.ConnectionHandler |