index 91fbbd0be6bb1181ad504475467085e6edd390ee..8f2a3ed5e4fa2759a3ae5769e7613174ce32d07c 100644 (file)
*
*
* IDENTIFICATION
- * $PostgreSQL: pgsql/src/backend/access/transam/xact.c,v 1.290 2010年05月13日 11:15:38 sriggs Exp $
+ * $PostgreSQL: pgsql/src/backend/access/transam/xact.c,v 1.291 2010年05月13日 11:39:30 sriggs Exp $
*
*-------------------------------------------------------------------------
*/
(void) XLogInsert(RM_XACT_ID, XLOG_XACT_ABORT, rdata);
+ /*
+ * Report the latest async abort LSN, so that the WAL writer knows to
+ * flush this abort. There's nothing to be gained by delaying this,
+ * since WALWriter may as well do this when it can. This is important
+ * with streaming replication because if we don't flush WAL regularly
+ * we will find that large aborts leave us with a long backlog for
+ * when commits occur after the abort, increasing our window of data
+ * loss should problems occur at that point.
+ */
+ if (!isSubXact)
+ XLogSetAsyncCommitLSN(XactLastRecEnd);
+
/*
* Mark the transaction aborted in clog. This is not absolutely necessary
* but we may as well do it while we are here; also, in the subxact case