LU-577 tests: FAIL replay-single test_70b rundbench load

Test 70b for replay-single assumes that lustre is mounted on
/mnt/lustre which is not the case for us. This patch passes
the proper MOUNT. The test also was not using the standard
DIR/tdir setup which had generated data files not being
cleaned up. Increased the sleep period to match dbench's
warm up period. This gives dbench a change to start up when
using many clients. Set the pdsh FANOUT environment variable
because by default pdsh launches in blocks of 32 nodes. This
way pdsh will lauch all node jobs at the same time

Signed-off-by: James Simmons <uja.ornl@gmail.com>
Signed-off-by: Yu Jian <yujian@whamcloud.com>
Change-Id: Iefdfe4bf47d322e782e138cf6ae17240b3004d8b
Reviewed-on: http://review.whamcloud.com/252
Tested-by: Hudson
Tested-by: Maloo <whamcloud.maloo@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

diff --git a/lustre/tests/replay-single.sh b/lustre/tests/replay-single.sh
index cac6ea4..857a22a 100755 (executable)
--- a/lustre/tests/replay-single.sh
+++ b/lustre/tests/replay-single.sh
@@ -1854,7 +1854,7 @@ test_70b () {
        local clients=${CLIENTS:-$HOSTNAME}
 
        zconf_mount_clients $clients $MOUNT
-       
+
        local duration=300
        [ "$SLOW" = "no" ] && duration=60
        # set duration to 900 because it takes some time to boot node
@@ -1863,20 +1863,20 @@ test_70b () {
        local cmd="rundbench 1 -t $duration"
        local pid=""
        do_nodesv $clients "set -x; MISSING_DBENCH_OK=$MISSING_DBENCH_OK \
-               PATH=:$PATH:$LUSTRE/utils:$LUSTRE/tests/:$DBENCH_LIB \
+               PATH=\$PATH:$LUSTRE/utils:$LUSTRE/tests/:$DBENCH_LIB \
                DBENCH_LIB=$DBENCH_LIB TESTSUITE=$TESTSUITE TESTNAME=$TESTNAME \
-               LCTL=$LCTL $cmd" &
+               MOUNT=$MOUNT DIR=$DIR/$tdir/\\\$(hostname) LCTL=$LCTL $cmd" &
        pid=$!
        log "Started rundbench load pid=$pid ..."
 
        # give rundbench a chance to start, bug 24118
-       sleep 2
+       sleep 12
        local elapsed=0
        local num_failovers=0
        local start_ts=$(date +%s)
        while [ $elapsed -lt $duration ]; do
-               if ! check_for_process $clients rundbench; then
-                       error_noexit "rundbench not found on some of $clients!"
+               if ! check_for_process $clients dbench; then
+                       error_noexit "dbench not found on some of $clients!"
                        killall_process $clients dbench
                        break
                fi

diff --git a/lustre/tests/test-framework.sh b/lustre/tests/test-framework.sh
index 0b1fc03..1ba7de5 100644 (file)
--- a/lustre/tests/test-framework.sh
+++ b/lustre/tests/test-framework.sh
@@ -2013,6 +2013,7 @@ do_nodes() {
     [ -z "$myPDSH" -o "$myPDSH" = "no_dsh" -o "$myPDSH" = "rsh" ] && \
         echo "cannot run remote command on $rnodes with $myPDSH" && return 128
 
+    export FANOUT=$(get_node_count "${rnodes//,/ }")
     if $VERBOSE; then
         echo "CMD: $rnodes $@" >&2
         $myPDSH $rnodes "$LCTL mark \"$@\"" > /dev/null 2>&1 || :