b=19086

i=rread,tappro

- remove call of df in backgroup from facet_failover()
- use client_df in fail() to wait for recovery completion
- replay_single:test_70b: use fail() instead of facet_failover
- replay_single:test_74: umount/mount all clients

diff --git a/lustre/tests/insanity.sh b/lustre/tests/insanity.sh
index 06b5ca3..4d4f770 100755 (executable)
--- a/lustre/tests/insanity.sh
+++ b/lustre/tests/insanity.sh
@@ -148,14 +148,10 @@ check_and_setup_lustre
 echo "Starting Test 17 at `date`"
 
 test_0() {
-    facet_failover $SINGLEMDS
-    echo "Waiting for df pid: $DFPID"
-    wait $DFPID || { echo "df returned $?" && return 1; }
+    fail $SINGLEMDS
 
     for i in $(seq $OSTCOUNT) ; do
-        facet_failover ost$i || return 4
-        echo "Waiting for df pid: $DFPID"
-        wait $DFPID || { echo "df returned $?" && return 3; }
+        fail ost$i
     done
     return 0
 }
@@ -215,8 +211,7 @@ test_3() {
     [ -z "$(mounted_lustre_filesystems)" ] && error "Lustre is not running"
     
     #MDS Portion
-    facet_failover $SINGLEMDS
-    wait $DFPID || echo df failed: $?
+    fail $SINGLEMDS
     #Check FS
 
     echo "Test Lustre stability after MDS failover"
@@ -419,11 +414,8 @@ test_7() {
     client_rm testfile
 
     #MDS Portion
-    facet_failover $SINGLEMDS
+    fail $SINGLEMDS
 
-    #Check FS
-    echo "Test Lustre stability after MDS failover"
-    wait $DFPID || echo "df on down clients fails " || return 1
     $PDSH $LIVE_CLIENT "ls -l $MOUNT"
     $PDSH $LIVE_CLIENT "rm -f $MOUNT/*_testfile"
 

diff --git a/lustre/tests/replay-single.sh b/lustre/tests/replay-single.sh
index b6222e0..65691bb 100755 (executable)
--- a/lustre/tests/replay-single.sh
+++ b/lustre/tests/replay-single.sh
@@ -1857,7 +1857,7 @@ test_70b () {
                # Increment the number of failovers
                NUM_FAILOVERS=$((NUM_FAILOVERS+1))
                log "$TESTNAME fail mds1 $NUM_FAILOVERS times"
-               facet_failover $SINGLEMDS
+               fail $SINGLEMDS
                CURRENT_TS=$(date +%s)
                ELAPSED=$((CURRENT_TS - START_TS))
        done
@@ -1917,13 +1917,13 @@ run_test 73c "open(O_CREAT), unlink, replay, reconnect at last_replay, close"
 # bug 18554
 test_74() {
     stop ost1
-    zconf_umount $(hostname) $MOUNT
-    fail $SINGLEMDS
-    zconf_mount $(hostname) $MOUNT
+    zconf_umount_clients $CLIENTS $MOUNT
+    facet_failover $SINGLEMDS
+    zconf_mount_clients $CLIENTS $MOUNT
     mount_facet ost1
     touch $DIR/$tfile || return 1
     rm $DIR/$tfile || return 2
-    df $MOUNT || error "df failed: $?"
+    client_df || error "df failed: $?"
     return 0
 }
 run_test 74 "Ensure applications don't fail waiting for OST reocvery"

diff --git a/lustre/tests/test-framework.sh b/lustre/tests/test-framework.sh
index a1b9f03..10f6f1f 100644 (file)
--- a/lustre/tests/test-framework.sh
+++ b/lustre/tests/test-framework.sh
@@ -1161,9 +1161,6 @@ facet_failover() {
     shutdown_facet $facet
     [ -n "$sleep_time" ] && sleep $sleep_time
     reboot_facet $facet
-    client_df &
-    DFPID=$!
-    echo "df pid is $DFPID"
     change_active $facet
     local TO=`facet_active_host $facet`
     echo "Failover $facet to $TO"
@@ -1219,7 +1216,7 @@ ost_evict_client() {
 
 fail() {
     facet_failover $* || error "failover: $?"
-    df $MOUNT || error "post-failover df: $?"
+    client_df || error "post-failover df: $?"
 }
 
 fail_nodf() {
@@ -1232,9 +1229,9 @@ fail_abort() {
     stop $facet
     change_active $facet
     mount_facet $facet -o abort_recovery
-    df $MOUNT || echo "first df failed: $?"
+    client_df || echo "first df failed: $?"
     sleep 1
-    df $MOUNT || error "post-failover df: $?"
+    client_df || error "post-failover df: $?"
 }
 
 do_lmc() {