Make sure that open hash table scans are cleaned up when bgwriter tries to

recover from elog(ERROR).  Problem was created by introduction of hash seq
search tracking awhile back, and affects all branches that have bgwriter;
in HEAD the disease has snuck into autovacuum and walwriter too.  (Not sure
that the latter two use hash_seq_search at the moment, but surely they might
someday.)  Per report from Sergey Koposov.

diff --git a/src/backend/postmaster/bgwriter.c b/src/backend/postmaster/bgwriter.c
index fefcbdf67b031b87e5f8e3aeb53f7435b0e81900..721933a814c07e07630a83761a2e24a2cace8b39 100644 (file)
--- a/src/backend/postmaster/bgwriter.c
+++ b/src/backend/postmaster/bgwriter.c
@@ -37,7 +37,7 @@
  *
  *
  * IDENTIFICATION
- *       $PostgreSQL: pgsql/src/backend/postmaster/bgwriter.c,v 1.21.2.2 2007/01/27 20:15:55 tgl Exp $
+ *       $PostgreSQL: pgsql/src/backend/postmaster/bgwriter.c,v 1.21.2.3 2007/09/11 17:15:48 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -260,6 +260,7 @@ BackgroundWriterMain(void)
                /* we needn't bother with the other ResourceOwnerRelease phases */
                AtEOXact_Buffers(false);
                AtEOXact_Files();
+               AtEOXact_HashTables(false);
 
                /* Warn any waiting backends that the checkpoint failed. */
                if (ckpt_active)