Re: [ndctl PATCH v2 3/3] ndctl, test: Add a unit test for the BTT checker

["Verma, Vishal L" <vishal.l.verma-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>]

On Tue, 2017-02-28 at 16:19 -0500, Jeff Moyer wrote:
> Vishal Verma <vishal.l.verma-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> writes:
> 
> > +# re-enable the BTT namespace, and do IO to it in an attempt to
> > +# verify it still comes up ok, and functions as expected
> > +post_repair_test()
> > +{
> > +	echo "${FUNCNAME[0]}: I/O to BTT namespace"
> > +	test -b /dev/$blockdev
> > +	dd if=/dev/zero of=/dev/$blockdev bs=$sector_size
> > count=$((size/sector_size)) > /dev/null 2>&1
> > +	dd if=/dev/$blockdev of=/dev/null bs=$sector_size
> > count=$((size/sector_size)) > /dev/null 2>&1
> 
> Maybe validate the data you got?

Good idea, I was just testing that the IOs don't fail, but I'll add
writing known data and validating it is read back the same.

> 
> > +	echo "done"
> > +}
> > +
> > +test_normal()
> > +{
> > +	echo "=== ${FUNCNAME[0]} ==="
> > +	# disable the namespace
> > +	$ndctl disable-namespace $dev
> > +	$ndctl check-namespace $dev
> 
> Shouldn't we ensure there are no errors reported by the check?

It should happen already. Check-namespace will return a non zero exit
code for any errors, and that will trigger the script's set -E error
trap.

> 
> > +	$ndctl enable-namespace $dev
> > +	post_repair_test
> > +}
> > +
> 
> [snip]
> 
> > +test_bitmap()
> > +{
> > +	echo "=== ${FUNCNAME[0]} ==="
> > +	reset && create
> > +	set_raw
> > +	# scribble over the last 4K of the map
> > +	rm -f /tmp/scribble
> > +	for (( i=0 ; i<512 ; i++ )); do
> > +		echo -n -e \\x1e\\x1e\\x00\\xc0\\x1e\\x1e\\x00\\xc0
> > >> /tmp/scribble
> > +	done
> > +	seek="$((raw_size/bs - (256*64/bs) - 2))"
> > +	echo "scribbling over map entries (offset = $seek blocks)"
> > +	dd if=/tmp/scribble of=/dev/$raw_bdev bs=$bs seek=$seek
> > +	rm -f /tmp/scribble
> > +	unset_raw
> > +	$ndctl disable-namespace $dev
> > +	$ndctl check-namespace $dev 2>&1 | grep "bitmap error"
> > +	# This is not repairable
> > +	reset && create
> > +}
> > +
> 
> Can you add a test to inject badblocks in the metadata?

I had been putting off badblock handling with BTT metadata, but you're
right, let me figure out a flow for how to handle/report this. Obviously
recovery is not possible at that point, but we should be able to provide
a coherent message.

> 
> Cheers,
> Jeff