Update of /home/cvsd/slony1/slony1-engine/tests/test1
In directory main.slony.info:/tmp/cvs-serv6415/test1

Modified Files:
      Tag: REL_2_0_STABLE
	generate_dml.sh 
Log Message:
Revise tests to NOT use poll_cluster to "wait until everything has
propagated according to sl_status" - instead, we use SYNC/WAIT FOR
EVENT, which submits an event against the origin, and then waits until
that specific event has propagated.

This means we're testing WAIT FOR EVENT a lot more stringently, and
allows tests to run a lot quicker in cases where it takes a while
for the nodes to decide they are instantaneously "in sync"


Index: generate_dml.sh
===================================================================
RCS file: /home/cvsd/slony1/slony1-engine/tests/test1/generate_dml.sh,v
retrieving revision 1.18.2.1
retrieving revision 1.18.2.2
diff -C2 -d -r1.18.2.1 -r1.18.2.2
*** generate_dml.sh	6 Apr 2009 22:33:05 -0000	1.18.2.1
--- generate_dml.sh	28 Apr 2009 21:48:19 -0000	1.18.2.2
***************
*** 71,75 ****
    eval port=\$PORT${originnode}
    generate_initdata
-   launch_poll
    status "run updateReloid() - equivalent to REPAIR NODE"
    $pgbindir/psql -h $host -p $port -d $db -U $user -c "select \"_${CLUSTER1}\".updateReloid(1, 0);" 1> $mktmp/reloidtest.log 2> $mktmp/reloidtest.log
--- 71,74 ----
***************
*** 97,100 ****
--- 96,100 ----
    fi
    status "completed cleanupEvent(${pint},${dellogs}) test"
+   wait_for_catchup
    status "done"
  }