| From: | Craig Ringer <craig(at)2ndquadrant(dot)com> |
|---|---|
| To: | "Davygora, Yuriy" <Yuriy(dot)Davygora(at)sulzer(dot)de> |
| Cc: | Lachezar Dobrev <l(dot)dobrev(at)gmail(dot)com>, Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com>, List <pgsql-jdbc(at)postgresql(dot)org> |
| Subject: | Re: JPA + Postgres = autocommit? |
| Date: | 2016-07-29 03:44:42 |
| Message-ID: | CAMsr+YERVmwD1fqgb6y=3KOF0zw03+1UNTj0DeCY2CFs4rTi-A@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-jdbc |
On 27 July 2016 at 16:50, Davygora, Yuriy <Yuriy(dot)Davygora(at)sulzer(dot)de> wrote:
>
>
> Hello Lachezar,
>
>
>
> thank you for your reply.
>
>
>
> First of all, let me list some of the things I had tried before I
> started digging into the source code of the Postgres JDBC driver and
> Eclipselink.
>
>
>
> - switching PGConnectionPoolDataSource.
>
> - both container managed and bean managed transactions.
>
> - REQUIRED and REQUIRES_NEW for container managed transactions.
>
>
>
> None of this worked.
>
IIRC with Glassfish back in the day I used org.postgresql.Driver and let
Glassfish's connection pooler wrap its own DataSource around it.
> I ruled out the possibility of a bug in the Postgres JDBC driver and in
> EclipseLink. Only Payara remains. As said, for some reason, setAutoCommit
> is never called, which is the job of the transaction manager.
>
Yeah, that's pretty suspect.
--
Craig Ringer http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Craig Ringer | 2016-07-29 04:08:41 | Re: Possible bug related to primary keys autogeneration |
| Previous Message | Vladimir Sitnikov | 2016-07-28 22:20:09 | Re: ResultSet.getClob() causing problems when used with JPA's @Lob |