+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+status=1 # failure is the default!
Please use './new btrfs' to generate new test template, which defines
variables like 'tmp' and 'here', these variables may not be used in the
test explicitly, but they could be used by some helper functions,
especially the "$tmp" var. (Or please don't remove them from template :)
Ok. May be its a good idea to have an _init() to set mandatory stuffs?
+MNT_OPT=$(echo $MOUNT_OPTIONS | cut -d" " -f2-)
I'm not sure about the purpose of this MNT_OPT, and it could be empty
and causes problems in test. Please see below.
Test case here need control of the device(s) used for the mount.
Thats awkward I missed no mount-option secnario. Will fix.
+ echo -e "\\ncheck_copy1_fsid\\n{" | tee -a $seqres.full
Hmm, I don't think the "{ }" block is necessary in .out file, IMHO it
makes the diff context harder to read when test fails. Just echo the
test name would be fine.
Ok. Will take that out.
+ _mount -o $MNT_OPT $DEV_BAD $SCRATCH_MNT 2>&1 | _filter_scratch_pool
When $MOUNT_OPTIONS is empty, this mount fails as
+mount: can't find /mnt/scratch in /etc/fstab
And I think we need to umount $SCRATCH_MNT here in case a buggy btrfs
module allowed the mount, as the next test assumes the device is not
mounted anywhere. Otherwise I see failures like:
+ERROR: /dev/loop1 is mounted
+mount failed
got it.
+
+ echo -e "good\\n}" | tee -a $seqres.full
+}
+
+# As in Linux kernel 4.16 if the primary is corrupted mount will fail.
+# Which might change in the long run.
+check_primary()
+{
+ local bytenr=65536
+ echo -e "\\ncheck_primary\\n{" | tee -a $seqres.full
+
+ wipe
+ _scratch_pool_mkfs "-mraid1 -draid1"
+ _reload_fs_module "btrfs"
+
+ #To corrupt a disk block, read in hex, write in dec
+ od -j$bytenr -N1 -An -x $DEV_BAD |\
+ dd status=none of=$DEV_BAD ibs=1 obs=1 seek=$bytenr count=1|\
+ tee -a $seqres.full
+ _mount -o $MNT_OPT,device=$DEV_GOOD $DEV_BAD $SCRATCH_MNT 2>&1 | _filter_scratch_pool
Same here, need to umount $SCRATCH_MNT.
Also, if a mount failure is expected (as it's recorded in the golden
image file), you need _filter_error_mount here, and _filter_scratch_pool
can be dropped, because _filter_error_mount will remove all
$SCRATCH_DEV/MNT from the mount output. For more details please see
commit 94d3a4f00cbd ("fstests: filter mount error message for EUCLEAN
and ESTALE").
I am fixing this in v2.
Thanks, Anand
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
