From: Greg Smith <gsmith@gregsmith.com>
Date: Mon, 15 Mar 2010 04:10:27 +0000 (-0400)
Subject: Change default to distinct results database
X-Git-Tag: v0.6.0~9
X-Git-Url: http://git.postgresql.org/gitweb/edit?a=commitdiff_plain;h=7309b208e98bbfb20b6f3309f1f69062d05d50c7;p=pgbench-tools.git

Change default to distinct results database
---

diff --git a/README b/README
index 36524f7..f233de6 100644
--- a/README
+++ b/README
@@ -1,16 +1,20 @@
 pgbench-tools Setup
 ===================
 
-* Create databases for your test and for the results
+* Create databases for your test and for the results::
+
+    createdb results
+    createdb pgbench
 
   *  Both databases can be the same, but there may be more shared_buffers
      cache churn in that case.  Some amount of cache disruption
      is unavoidable unless the result database is remote, because
-     of the OS cache.
+     of the OS cache.  The recommended and default configuration
+     is to have a pgbench database and a results database.
 
 * Initialize the results database by executing::
 
-    psql -f init/resultdb.sql -d pgbench
+    psql -f init/resultdb.sql -d results
 
   Make sure to reference the correct database.
   This will create a default test set entry with a blank description.
@@ -32,7 +36,7 @@ Results
 
 * You can check results even as the test is running with::
 
-    psql [resultdb] -f report.sql
+    psql -d results -f report.sql
 
 * Other useful reports you can run include:
    * summary.sql
diff --git a/config b/config
index 55a9cc8..bc17903 100755
--- a/config
+++ b/config
@@ -11,13 +11,13 @@ TESTDIR="tests"
 
 TESTHOST=localhost
 TESTUSER=postgres
-TESTDB=pgbench
 TESTPORT=5432
+TESTDB=pgbench
 
 RESULTHOST="$TESTHOST"
 RESULTUSER="$TESTUSER"
-RESULTDB="$TESTDB"
 RESULTPORT="$TESTPORT"
+RESULTDB=results
 
 SCALES="1 10 100 1000"
 SCRIPT="select.sql"