Regression test portability: ec.t

This test unmount/remount the filesystem to invalidate cache,
but this leads to timing problems on NetBSD. We can work them
around without sleeping by remounting on another mount point.

BUG: 1129939
Change-Id: I10b3183e5e715053de162a6980af188710b607bb
Signed-off-by: Emmanuel Dreyfus <manu@netbsd.org>
Reviewed-on: http://review.gluster.org/9285
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
This commit is contained in:
Emmanuel Dreyfus 2014-12-16 17:06:20 +01:00 committed by Vijay Bellur
parent e398f99d9a
commit 3abee447da

View File

@ -222,12 +222,12 @@ sleep 2
# Unmount/remount so that create/write and truncate don't see cached data.
TEST umount $M0
TEST $GFS -s $H0 --volfile-id $V0 $M0
TEST $GFS -s $H0 --volfile-id $V0 $M1
EXPECT_WITHIN $CHILD_UP_TIMEOUT "8" ec_child_up_count $V0 0
# Test create/write and truncate *before* the bricks are brought back.
TEST check_create_write $M0
TEST check_truncate $M0
TEST check_create_write $M1
TEST check_truncate $M1
# Restart the bricks and allow repair to occur.
TEST $CLI volume start $V0 force
@ -235,7 +235,7 @@ EXPECT_WITHIN $PROCESS_UP_TIMEOUT 'Started' volinfo_field $V0 'Status'
EXPECT_WITHIN $CHILD_UP_TIMEOUT "10" ec_child_up_count $V0 0
# Unmount/remount again, same reason as before.
TEST umount $M0
TEST umount $M1
TEST $GFS -s $H0 --volfile-id $V0 $M0
EXPECT_WITHIN $CHILD_UP_TIMEOUT "10" ec_child_up_count $V0 0