Re: [Qemu-devel] [PATCH v2 1/3] iotests: Add 241 to test NBD on unaligned images

[Eric Blake]

On 3/28/19 3:20 AM, Vladimir Sementsov-Ogievskiy wrote:
> 28.03.2019 1:39, Eric Blake wrote:
>> Add a test for the NBD client workaround in the previous patch.  It's
>> not really feasible for an iotest to assume a specific tracing engine,
>> so we can't really probe for the new
>> trace_nbd_parse_blockstatus_compliance to see if the server was fixed
>> vs. whether the client just worked around the server (other than by
>> rearranging order between code patches and this test). But having a
>> successful exchange sure beats the previous state of an error message.
>>

>> +seq="$(basename $0)"
>> +echo "QA output created by $seq"
>> +
>> +status=1 # failure is the default!
>> +
>> +nbd_unix_socket=$TEST_DIR/test_qemu_nbd_socket
>> +rm -f "${TEST_DIR}/qemu-nbd.pid"
> 
> hmm, strange that we need to remove something from test directory at start.

Well, until we follow through with our thread of implementing per-test
scratch directories for iotests, it proved invaluable to me during
testing (as a failed test does not properly clean up after itself).


>> +# can't use _make_test_img, because qemu-img rounds image size up,
>> +# and because we want to use Unix socket rather than TCP port. Likewise,
>> +# we have to redirect TEST_IMG to our server.
>> +# This tests that we can deal with the hole at the end of an unaligned
>> +# raw file (either because the server doesn't advertise alignment too
>> +# large, or because the client ignores the server's noncompliance).
>> +printf %01000d 0 > "$TEST_IMG_FILE"
>> +nbd_server_start_unix_socket -f $IMGFMT -e 42 -x '' "$TEST_IMG_FILE"
> 
> why do we need -e 42? we don't have more than one client simultaneously..

Matches what other tests use. You are right that I could get away
without it, though.

> and -x '' is the default anyway

Ditto.

> 
>> +TEST_IMG="nbd:unix:$nbd_unix_socket"
>> +
>> +$QEMU_IMG map --output=json "$TEST_IMG" | _filter_qemu_img_map
> 
> is filter needed? seems like not

I'd rather keep the filter (makes it easier to copy-and-paste between
other tests, especially if it does matter down the road)

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org

signature.asc
Description: OpenPGP digital signature