LU-5075 test: keep LFSCK fail_loc until recovery completed

In sanity-lfsck test_8, the old script may over wrote the fail_loc
before the recovery completed, that will cause the paused LFSCK to
be restarted automatically, then the subsequent manual command for
starting the LFSCK will get -EALREADY. That is unexpected.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I8500ecdccf354e83ec70be8dea88943cafa47d81
Reviewed-on: http://review.whamcloud.com/11288
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

diff --git a/lustre/tests/sanity-lfsck.sh b/lustre/tests/sanity-lfsck.sh
index 18c6bbb..ea3df21 100644 (file)
--- a/lustre/tests/sanity-lfsck.sh
+++ b/lustre/tests/sanity-lfsck.sh
@@ -729,6 +729,21 @@ test_8()
        start $SINGLEMDS $MDT_DEVNAME $MOUNT_OPTS_SCRUB > /dev/null ||
                error "(14) Fail to start MDS!"
 
        start $SINGLEMDS $MDT_DEVNAME $MOUNT_OPTS_SCRUB > /dev/null ||
                error "(14) Fail to start MDS!"
 

+       local timeout=$(max_recovery_time)
+       local timer=0
+
+       while [ $timer -lt $timeout ]; do
+               STATUS=$(do_facet $SINGLEMDS "$LCTL get_param -n \
+                       mdt.${MDT_DEV}.recovery_status |
+                       awk '/^status/ { print \\\$2 }'")
+               [ "$STATUS" != "RECOVERING" ] && break;
+               sleep 1
+               timer=$((timer + 1))
+       done
+
+       [ $timer != $timeout ] ||
+               error "(14.1) recovery timeout"
+

        STATUS=$($SHOW_NAMESPACE | awk '/^status/ { print $2 }')
        [ "$STATUS" == "crashed" ] ||
                error "(15) Expect 'crashed', but got '$STATUS'"
        STATUS=$($SHOW_NAMESPACE | awk '/^status/ { print $2 }')
        [ "$STATUS" == "crashed" ] ||
                error "(15) Expect 'crashed', but got '$STATUS'"


@@ -751,6 +766,19 @@ test_8()
        start $SINGLEMDS $MDT_DEVNAME $MOUNT_OPTS_SCRUB > /dev/null ||
                error "(19) Fail to start MDS!"
 
        start $SINGLEMDS $MDT_DEVNAME $MOUNT_OPTS_SCRUB > /dev/null ||
                error "(19) Fail to start MDS!"
 

+       timer=0
+       while [ $timer -lt $timeout ]; do
+               STATUS=$(do_facet $SINGLEMDS "$LCTL get_param -n \
+                       mdt.${MDT_DEV}.recovery_status |
+                       awk '/^status/ { print \\\$2 }'")
+               [ "$STATUS" != "RECOVERING" ] && break;
+               sleep 1
+               timer=$((timer + 1))
+       done
+
+       [ $timer != $timeout ] ||
+               error "(19.1) recovery timeout"
+

        STATUS=$($SHOW_NAMESPACE | awk '/^status/ { print $2 }')
        [ "$STATUS" == "paused" ] ||
                error "(20) Expect 'paused', but got '$STATUS'"
        STATUS=$($SHOW_NAMESPACE | awk '/^status/ { print $2 }')
        [ "$STATUS" == "paused" ] ||
                error "(20) Expect 'paused', but got '$STATUS'"