Kevin Grittner <[email protected]> wrote:
> Tom Lane <[email protected]> wrote:
>> A comment seems essential here, because as written anybody would
>> think the test for a snapshot is a bug.
>
> Good point.
I propose to apply the attached to master and back-patch to 9.3,
and follow that with a patch (for master only) along the lines
suggested by Andres. Since *that* change is more invasive and
changes existing behavior I will submit it to the open CF for
review. Objections?
--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
diff --git a/src/bin/pg_dump/pg_dump.c b/src/bin/pg_dump/pg_dump.c
index c3ebb3a..6d57fe0 100644
--- a/src/bin/pg_dump/pg_dump.c
+++ b/src/bin/pg_dump/pg_dump.c
@@ -1011,7 +1011,17 @@ setup_connection(Archive *AH, DumpOptions *dopt, const char *dumpencoding,
ExecuteSqlStatement(AH, "BEGIN");
if (AH->remoteVersion >= 90100)
{
- if (dopt->serializable_deferrable)
+ if (dopt->serializable_deferrable &&
+ /*
+ * To support the combination of this option with the jobs option
+ * we use REPEATABLE READ for the worker connections that are
+ * passed a snapshot. As long as the snapshot is acquired in a
+ * SERIALIZABLE, READ ONLY, DEFERRABLE transaction, its use within
+ * a REPEATABLE READ transaction provides the appropriate
+ * integrity guarantees. This is a kluge, but safe for
+ * back-patching.
+ */
+ AH->sync_snapshot_id == NULL)
ExecuteSqlStatement(AH,
"SET TRANSACTION ISOLATION LEVEL "
"SERIALIZABLE, READ ONLY, DEFERRABLE");
--
Sent via pgsql-hackers mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
