Okay, attached is a patch that moves the alter table regression test

into a separate parallel group. I can't confirm that this fixes the
problem for me (since it only occurs about one in ten or one in twenty
runs), but I think it's a good bet that it does...

Neil Conway

src/test/regress/parallel_schedule

@@ -68,10 +68,10 @@ test: misc
 # ----------
 # The fifth group of parallel test
 # ----------
-test: select_views alter_table portals_p2 rules foreign_key cluster
+test: select_views portals_p2 rules foreign_key cluster
 
 # ----------
 # The sixth group of parallel test
 # ----------
 # "plpgsql" cannot run concurrently with "rules"
-test: limit plpgsql copy2 temp domain rangefuncs prepare without_oid conversion truncate
+test: limit plpgsql copy2 temp domain rangefuncs prepare without_oid conversion truncate alter_table

src/test/regress/serial_schedule

-# $Header: /cvsroot/pgsql/src/test/regress/serial_schedule,v 1.17 2002/08/27 04:55:12 tgl Exp $
+# $Header: /cvsroot/pgsql/src/test/regress/serial_schedule,v 1.18 2002/09/02 05:55:43 momjian Exp $
 # This should probably be in an order similar to parallel_schedule.
 test: boolean
 test: char
@@ -75,7 +75,6 @@ test: hash_index
 test: privileges
 test: misc
 test: select_views
-test: alter_table
 test: portals_p2
 test: rules
 test: foreign_key
@@ -90,3 +89,4 @@ test: prepare
 test: without_oid
 test: conversion
 test: truncate
+test: alter_table