[v3,2/4] iotests: Switch nbd tests to use Unix rather than TCP
diff mbox series

Message ID 20191114213415.23499-3-eblake@redhat.com
State New
Headers show
Series
  • tests: More iotest 223 improvements
Related show

Commit Message

Eric Blake Nov. 14, 2019, 9:34 p.m. UTC
Up to now, all it took to cause a lot of iotest failures was to have a
background process such as 'nbdkit -p 10810 null' running, because we
hard-coded the TCP port.  Switching to a Unix socket eliminates this
contention.  We still have TCP coverage in test 233, and that test is
more careful to not pick a hard-coded port.

Signed-off-by: Eric Blake <eblake@redhat.com>
---
 tests/qemu-iotests/common.filter | 6 ++++--
 tests/qemu-iotests/common.rc     | 8 ++++----
 2 files changed, 8 insertions(+), 6 deletions(-)

Comments

Max Reitz Nov. 18, 2019, 5:29 p.m. UTC | #1
On 14.11.19 22:34, Eric Blake wrote:
> Up to now, all it took to cause a lot of iotest failures was to have a
> background process such as 'nbdkit -p 10810 null' running, because we
> hard-coded the TCP port.  Switching to a Unix socket eliminates this
> contention.  We still have TCP coverage in test 233, and that test is
> more careful to not pick a hard-coded port.

For me, all it took was to run qcow2 and nbd tests in parallel (some
qcow2 tests create nbd servers, too), so this is great.

> Signed-off-by: Eric Blake <eblake@redhat.com>
> ---
>  tests/qemu-iotests/common.filter | 6 ++++--
>  tests/qemu-iotests/common.rc     | 8 ++++----
>  2 files changed, 8 insertions(+), 6 deletions(-)
> 
> diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
> index f870e00e4421..5367deea398e 100644
> --- a/tests/qemu-iotests/common.filter
> +++ b/tests/qemu-iotests/common.filter
> @@ -127,7 +127,8 @@ _filter_img_create()
>          -e "s#$TEST_DIR#TEST_DIR#g" \
>          -e "s#$SOCK_DIR#SOCK_DIR#g" \
>          -e "s#$IMGFMT#IMGFMT#g" \
> -        -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \
> +        -e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
> +        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \

Why the second question mark?  I thought the ? after the /// was mandatory.

>          -e "s# encryption=off##g" \
>          -e "s# cluster_size=[0-9]\\+##g" \
>          -e "s# table_size=[0-9]\\+##g" \
> @@ -164,7 +165,8 @@ _filter_img_info()
>          -e "s#$TEST_DIR#TEST_DIR#g" \
>          -e "s#$SOCK_DIR#SOCK_DIR#g" \
>          -e "s#$IMGFMT#IMGFMT#g" \
> -        -e 's#nbd://127.0.0.1:10810$#TEST_DIR/t.IMGFMT#g' \
> +        -e 's#nbd://127.0.0.1:[0-9]\\+$#TEST_DIR/t.IMGFMT#g' \
> +        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
>          -e 's#json.*vdisk-id.*vxhs"}}#TEST_DIR/t.IMGFMT#' \
>          -e "/encrypted: yes/d" \
>          -e "/cluster_size: [0-9]\\+/d" \
> diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
> index fa7bae24226a..f772dcb67322 100644
> --- a/tests/qemu-iotests/common.rc
> +++ b/tests/qemu-iotests/common.rc
> @@ -217,7 +217,7 @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then
>          TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
>      elif [ "$IMGPROTO" = "nbd" ]; then
>          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
> -        TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
> +        TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT"

Maybe nbd.$IMGFMT?

>      elif [ "$IMGPROTO" = "ssh" ]; then
>          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>          TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
> @@ -233,7 +233,7 @@ else
>          TEST_IMG=$TEST_DIR/t.$IMGFMT
>      elif [ "$IMGPROTO" = "nbd" ]; then
>          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
> -        TEST_IMG="nbd:127.0.0.1:10810"
> +        TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"

Shouldn’t this be $IMGFMT, too (instead of nbd)?  (Or maybe nbd.$IMGFMT)

Max

>      elif [ "$IMGPROTO" = "ssh" ]; then
>          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>          REMOTE_TEST_DIR="ssh://\\($USER@\\)\\?127.0.0.1\\(:[0-9]\\+\\)\\?$TEST_DIR"
> @@ -293,7 +293,7 @@ _stop_nbd_server()
>          local QEMU_NBD_PID
>          read QEMU_NBD_PID < "${QEMU_TEST_DIR}/qemu-nbd.pid"
>          kill ${QEMU_NBD_PID}
> -        rm -f "${QEMU_TEST_DIR}/qemu-nbd.pid"
> +        rm -f "${QEMU_TEST_DIR}/qemu-nbd.pid" "$SOCK_DIR/nbd"
>      fi
>  }
> 
> @@ -353,7 +353,7 @@ _make_test_img()
>      if [ $IMGPROTO = "nbd" ]; then
>          # Pass a sufficiently high number to -e that should be enough for all
>          # tests
> -        eval "$QEMU_NBD -v -t -b 127.0.0.1 -p 10810 -f $IMGFMT -e 42 -x '' $TEST_IMG_FILE >/dev/null &"
> +        eval "$QEMU_NBD -v -t -k '$SOCK_DIR/nbd' -f $IMGFMT -e 42 -x '' $TEST_IMG_FILE >/dev/null &"
>          sleep 1 # FIXME: qemu-nbd needs to be listening before we continue
>      fi
>
Eric Blake Nov. 18, 2019, 5:42 p.m. UTC | #2
On 11/18/19 11:29 AM, Max Reitz wrote:
> On 14.11.19 22:34, Eric Blake wrote:
>> Up to now, all it took to cause a lot of iotest failures was to have a
>> background process such as 'nbdkit -p 10810 null' running, because we
>> hard-coded the TCP port.  Switching to a Unix socket eliminates this
>> contention.  We still have TCP coverage in test 233, and that test is
>> more careful to not pick a hard-coded port.
> 
> For me, all it took was to run qcow2 and nbd tests in parallel (some
> qcow2 tests create nbd servers, too), so this is great.
> 
>> Signed-off-by: Eric Blake <eblake@redhat.com>
>> ---
>>   tests/qemu-iotests/common.filter | 6 ++++--
>>   tests/qemu-iotests/common.rc     | 8 ++++----
>>   2 files changed, 8 insertions(+), 6 deletions(-)
>>
>> diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
>> index f870e00e4421..5367deea398e 100644
>> --- a/tests/qemu-iotests/common.filter
>> +++ b/tests/qemu-iotests/common.filter
>> @@ -127,7 +127,8 @@ _filter_img_create()
>>           -e "s#$TEST_DIR#TEST_DIR#g" \
>>           -e "s#$SOCK_DIR#SOCK_DIR#g" \
>>           -e "s#$IMGFMT#IMGFMT#g" \
>> -        -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \
>> +        -e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
>> +        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
> 
> Why the second question mark?  I thought the ? after the /// was mandatory.

Some of our code outputs:

nbd+unix://?socket=...

when there is no export name, while other outputs:

nbd+unix:///?socket=...

When there IS an export name, it outputs

nbd+unix:///name?socket=...

So the regex is matching 2 or 3 / (using \? to make the third optional), 
then a mandatory ?.


>> +++ b/tests/qemu-iotests/common.rc
>> @@ -217,7 +217,7 @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then
>>           TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
>>       elif [ "$IMGPROTO" = "nbd" ]; then
>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>> -        TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
>> +        TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT"
> 
> Maybe nbd.$IMGFMT?

At first glance, it seems reasonable.  But reading further,

> 
>>       elif [ "$IMGPROTO" = "ssh" ]; then
>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>           TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
>> @@ -233,7 +233,7 @@ else
>>           TEST_IMG=$TEST_DIR/t.$IMGFMT
>>       elif [ "$IMGPROTO" = "nbd" ]; then
>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>> -        TEST_IMG="nbd:127.0.0.1:10810"
>> +        TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"
> 
> Shouldn’t this be $IMGFMT, too (instead of nbd)?  (Or maybe nbd.$IMGFMT)

Now I'm starting to wonder.  With NBD and non-raw, there are two places 
to do the image format:

qcow2 file -> qemu-nbd -f qcow2 -> raw bytes over NBD -> qemu client -f 
raw -> guest  (our typical usage)

qcow2 file -> qemu-nbd -f raw -> qcow2 bytes over NBD -> qemu client -f 
qcow2 -> guest  (limited use, since NBD does not [yet] have resize support]

so naming the socket $SOCK_DIR/nbd.qcow2 when the socket carries raw 
data (our typical use) seems awkward.  But then again, running './check 
-qcow2 -nbd' shows that we seldom test qcow2 format over nbd protocol 
(precisely because nbd does not yet have resize).

If anything, I'm inclined to use $SOCK_DIR/nbd.raw to indicate that the 
NBD client sees raw format, regardless of the format in use by the 
server, to leave the door open for $SOCK_DIR/nbd.qcow2 when we finally 
are happy to test qcow2 format over NBD.

Or stick to just $SOCK_DIR/nbd hard-coded everywhere, and quit trying to 
use $IMGFMT in the socket name, to make all the usage consistent.
Eric Blake Nov. 18, 2019, 5:57 p.m. UTC | #3
On 11/18/19 11:42 AM, Eric Blake wrote:

>>> +++ b/tests/qemu-iotests/common.filter
>>> @@ -127,7 +127,8 @@ _filter_img_create()
>>>           -e "s#$TEST_DIR#TEST_DIR#g" \
>>>           -e "s#$SOCK_DIR#SOCK_DIR#g" \
>>>           -e "s#$IMGFMT#IMGFMT#g" \
>>> -        -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \
>>> +        -e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
>>> +        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
>>
>> Why the second question mark?  I thought the ? after the /// was 
>> mandatory.
> 
> Some of our code outputs:
> 
> nbd+unix://?socket=...
> 
> when there is no export name, while other outputs:
> 
> nbd+unix:///?socket=...

Re-reading 
https://github.com/NetworkBlockDevice/nbd/blob/master/doc/uri.md and RFC 
3986, I see that both forms appear to be valid URIs (both have empty 
authority, the first has empty path and the second has path '/', but NBD 
says leading / in path is stripped to form the export name).  However, 
the NBD document does not mention the 2-slash form with no URI authority 
or export name; perhaps we should amend that document to make it obvious 
that it is indeed valid.

> 
> When there IS an export name, it outputs
> 
> nbd+unix:///name?socket=...

While in this case, the 3-slash form is essential - with only two 
slashes, the URI 'nbd+unix://name?socket' would be trying to access a 
URI authority of 'name' with an empty path, rather than the intended 
empty authority and path of '/name' which gets translated to the NBD 
export 'name'.

> 
> So the regex is matching 2 or 3 / (using \? to make the third optional), 
> then a mandatory ?.
>
Max Reitz Nov. 18, 2019, 6:44 p.m. UTC | #4
On 18.11.19 18:42, Eric Blake wrote:
> On 11/18/19 11:29 AM, Max Reitz wrote:
>> On 14.11.19 22:34, Eric Blake wrote:
>>> Up to now, all it took to cause a lot of iotest failures was to have a
>>> background process such as 'nbdkit -p 10810 null' running, because we
>>> hard-coded the TCP port.  Switching to a Unix socket eliminates this
>>> contention.  We still have TCP coverage in test 233, and that test is
>>> more careful to not pick a hard-coded port.
>>
>> For me, all it took was to run qcow2 and nbd tests in parallel (some
>> qcow2 tests create nbd servers, too), so this is great.
>>
>>> Signed-off-by: Eric Blake <eblake@redhat.com>
>>> ---
>>>   tests/qemu-iotests/common.filter | 6 ++++--
>>>   tests/qemu-iotests/common.rc     | 8 ++++----
>>>   2 files changed, 8 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/tests/qemu-iotests/common.filter
>>> b/tests/qemu-iotests/common.filter
>>> index f870e00e4421..5367deea398e 100644
>>> --- a/tests/qemu-iotests/common.filter
>>> +++ b/tests/qemu-iotests/common.filter
>>> @@ -127,7 +127,8 @@ _filter_img_create()
>>>           -e "s#$TEST_DIR#TEST_DIR#g" \
>>>           -e "s#$SOCK_DIR#SOCK_DIR#g" \
>>>           -e "s#$IMGFMT#IMGFMT#g" \
>>> -        -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \
>>> +        -e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
>>> +        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
>>
>> Why the second question mark?  I thought the ? after the /// was
>> mandatory.
> 
> Some of our code outputs:
> 
> nbd+unix://?socket=...
> 
> when there is no export name, while other outputs:
> 
> nbd+unix:///?socket=...
> 
> When there IS an export name, it outputs
> 
> nbd+unix:///name?socket=...
> 
> So the regex is matching 2 or 3 / (using \? to make the third optional),
> then a mandatory ?.

Ah, I mixed up the escaping, as I often do when dealing with regexes.

>>> +++ b/tests/qemu-iotests/common.rc
>>> @@ -217,7 +217,7 @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then
>>>           TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
>>>       elif [ "$IMGPROTO" = "nbd" ]; then
>>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>> -       
>>> TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
>>> +       
>>> TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT"
>>>
>>
>> Maybe nbd.$IMGFMT?
> 
> At first glance, it seems reasonable.  But reading further,
> 
>>
>>>       elif [ "$IMGPROTO" = "ssh" ]; then
>>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>>          
>>> TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
>>>
>>> @@ -233,7 +233,7 @@ else
>>>           TEST_IMG=$TEST_DIR/t.$IMGFMT
>>>       elif [ "$IMGPROTO" = "nbd" ]; then
>>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>> -        TEST_IMG="nbd:127.0.0.1:10810"
>>> +        TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"
>>
>> Shouldn’t this be $IMGFMT, too (instead of nbd)?  (Or maybe nbd.$IMGFMT)
> 
> Now I'm starting to wonder.  With NBD and non-raw, there are two places
> to do the image format:
> 
> qcow2 file -> qemu-nbd -f qcow2 -> raw bytes over NBD -> qemu client -f
> raw -> guest  (our typical usage)
> 
> qcow2 file -> qemu-nbd -f raw -> qcow2 bytes over NBD -> qemu client -f
> qcow2 -> guest  (limited use, since NBD does not [yet] have resize support]
> 
> so naming the socket $SOCK_DIR/nbd.qcow2 when the socket carries raw
> data (our typical use) seems awkward.  But then again, running './check
> -qcow2 -nbd' shows that we seldom test qcow2 format over nbd protocol
> (precisely because nbd does not yet have resize).

Yes, the socket would carry qcow2 data.  Which doesn’t work (and I don’t
see a reason why we’d want it), but, well.

> If anything, I'm inclined to use $SOCK_DIR/nbd.raw to indicate that the
> NBD client sees raw format, regardless of the format in use by the
> server, to leave the door open for $SOCK_DIR/nbd.qcow2 when we finally
> are happy to test qcow2 format over NBD.

Sure, works for me.

> Or stick to just $SOCK_DIR/nbd hard-coded everywhere, and quit trying to
> use $IMGFMT in the socket name, to make all the usage consistent.

Works for me, too.

Max
Eric Blake Nov. 18, 2019, 10:18 p.m. UTC | #5
On 11/18/19 11:42 AM, Eric Blake wrote:

>>> -        
>>> TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
>>> +        
>>> TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT" 
>>>
>>
>> Maybe nbd.$IMGFMT?
> 
> At first glance, it seems reasonable.  But reading further,
> 
>>
>>>       elif [ "$IMGPROTO" = "ssh" ]; then
>>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>>           
>>> TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE" 
>>>
>>> @@ -233,7 +233,7 @@ else
>>>           TEST_IMG=$TEST_DIR/t.$IMGFMT
>>>       elif [ "$IMGPROTO" = "nbd" ]; then
>>>           TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
>>> -        TEST_IMG="nbd:127.0.0.1:10810"
>>> +        TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"
>>
>> Shouldn’t this be $IMGFMT, too (instead of nbd)?  (Or maybe nbd.$IMGFMT)

> 
> If anything, I'm inclined to use $SOCK_DIR/nbd.raw to indicate that the 
> NBD client sees raw format, regardless of the format in use by the 
> server, to leave the door open for $SOCK_DIR/nbd.qcow2 when we finally 
> are happy to test qcow2 format over NBD.

Naming the socket $SOCK_DIR/nbd.raw means that filters tend to rename it 
to $SOCK_DIR/nbd.IMGFMT before my attempt to rename everything to 
TEST_DIR/t.IMGFMT.  So I'm now leaning towards just naming the socket 
$SOCK_DIR/nbd and leave it at that.

> 
> Or stick to just $SOCK_DIR/nbd hard-coded everywhere, and quit trying to 
> use $IMGFMT in the socket name, to make all the usage consistent.
>
Eric Blake Nov. 18, 2019, 10:36 p.m. UTC | #6
On 11/18/19 4:18 PM, Eric Blake wrote:

>>
>> If anything, I'm inclined to use $SOCK_DIR/nbd.raw to indicate that 
>> the NBD client sees raw format, regardless of the format in use by the 
>> server, to leave the door open for $SOCK_DIR/nbd.qcow2 when we finally 
>> are happy to test qcow2 format over NBD.
> 
> Naming the socket $SOCK_DIR/nbd.raw means that filters tend to rename it 
> to $SOCK_DIR/nbd.IMGFMT before my attempt to rename everything to 
> TEST_DIR/t.IMGFMT.  So I'm now leaning towards just naming the socket 
> $SOCK_DIR/nbd and leave it at that.
> 
>>
>> Or stick to just $SOCK_DIR/nbd hard-coded everywhere, and quit trying 
>> to use $IMGFMT in the socket name, to make all the usage consistent.

In order to get my NBD 4.2-rc2 pull request out, I'll be squashing this 
in (having tested that my usual iotest configurations still pass)::

diff --git i/tests/qemu-iotests/common.rc w/tests/qemu-iotests/common.rc
index f772dcb67322..0cc8acc9edd2 100644
--- i/tests/qemu-iotests/common.rc
+++ w/tests/qemu-iotests/common.rc
@@ -217,7 +217,7 @@ if [ "$IMGOPTSSYNTAX" = "true" ]; then
          TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
      elif [ "$IMGPROTO" = "nbd" ]; then
          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
- 
TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT"
+ 
TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/nbd"
      elif [ "$IMGPROTO" = "ssh" ]; then
          TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
 
TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
@@ -349,7 +349,10 @@ _make_test_img()
       fi
      ) | _filter_img_create

-    # Start an NBD server on the image file, which is what we'll be 
talking to
+    # Start an NBD server on the image file, which is what we'll be 
talking to.
+    # Once NBD gains resize support, we may also want to use -f raw at the
+    # server and interpret format over NBD, but for now, the format is
+    # interpreted at the server and raw data sent over NBD.
      if [ $IMGPROTO = "nbd" ]; then
          # Pass a sufficiently high number to -e that should be enough 
for all
          # tests

Patch
diff mbox series

diff --git a/tests/qemu-iotests/common.filter b/tests/qemu-iotests/common.filter
index f870e00e4421..5367deea398e 100644
--- a/tests/qemu-iotests/common.filter
+++ b/tests/qemu-iotests/common.filter
@@ -127,7 +127,8 @@  _filter_img_create()
         -e "s#$TEST_DIR#TEST_DIR#g" \
         -e "s#$SOCK_DIR#SOCK_DIR#g" \
         -e "s#$IMGFMT#IMGFMT#g" \
-        -e 's#nbd:127.0.0.1:10810#TEST_DIR/t.IMGFMT#g' \
+        -e 's#nbd:127.0.0.1:[0-9]\\+#TEST_DIR/t.IMGFMT#g' \
+        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
         -e "s# encryption=off##g" \
         -e "s# cluster_size=[0-9]\\+##g" \
         -e "s# table_size=[0-9]\\+##g" \
@@ -164,7 +165,8 @@  _filter_img_info()
         -e "s#$TEST_DIR#TEST_DIR#g" \
         -e "s#$SOCK_DIR#SOCK_DIR#g" \
         -e "s#$IMGFMT#IMGFMT#g" \
-        -e 's#nbd://127.0.0.1:10810$#TEST_DIR/t.IMGFMT#g' \
+        -e 's#nbd://127.0.0.1:[0-9]\\+$#TEST_DIR/t.IMGFMT#g' \
+        -e 's#nbd+unix:///\??socket=SOCK_DIR/nbd#TEST_DIR/t.IMGFMT#g' \
         -e 's#json.*vdisk-id.*vxhs"}}#TEST_DIR/t.IMGFMT#' \
         -e "/encrypted: yes/d" \
         -e "/cluster_size: [0-9]\\+/d" \
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index fa7bae24226a..f772dcb67322 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -217,7 +217,7 @@  if [ "$IMGOPTSSYNTAX" = "true" ]; then
         TEST_IMG="$DRIVER,file.filename=$TEST_DIR/t.$IMGFMT"
     elif [ "$IMGPROTO" = "nbd" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
-        TEST_IMG="$DRIVER,file.driver=nbd,file.host=127.0.0.1,file.port=10810"
+        TEST_IMG="$DRIVER,file.driver=nbd,file.type=unix,file.path=$SOCKDIR/$IMGFMT"
     elif [ "$IMGPROTO" = "ssh" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
         TEST_IMG="$DRIVER,file.driver=ssh,file.host=127.0.0.1,file.path=$TEST_IMG_FILE"
@@ -233,7 +233,7 @@  else
         TEST_IMG=$TEST_DIR/t.$IMGFMT
     elif [ "$IMGPROTO" = "nbd" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
-        TEST_IMG="nbd:127.0.0.1:10810"
+        TEST_IMG="nbd+unix:///?socket=$SOCK_DIR/nbd"
     elif [ "$IMGPROTO" = "ssh" ]; then
         TEST_IMG_FILE=$TEST_DIR/t.$IMGFMT
         REMOTE_TEST_DIR="ssh://\\($USER@\\)\\?127.0.0.1\\(:[0-9]\\+\\)\\?$TEST_DIR"
@@ -293,7 +293,7 @@  _stop_nbd_server()
         local QEMU_NBD_PID
         read QEMU_NBD_PID < "${QEMU_TEST_DIR}/qemu-nbd.pid"
         kill ${QEMU_NBD_PID}
-        rm -f "${QEMU_TEST_DIR}/qemu-nbd.pid"
+        rm -f "${QEMU_TEST_DIR}/qemu-nbd.pid" "$SOCK_DIR/nbd"
     fi
 }

@@ -353,7 +353,7 @@  _make_test_img()
     if [ $IMGPROTO = "nbd" ]; then
         # Pass a sufficiently high number to -e that should be enough for all
         # tests
-        eval "$QEMU_NBD -v -t -b 127.0.0.1 -p 10810 -f $IMGFMT -e 42 -x '' $TEST_IMG_FILE >/dev/null &"
+        eval "$QEMU_NBD -v -t -k '$SOCK_DIR/nbd' -f $IMGFMT -e 42 -x '' $TEST_IMG_FILE >/dev/null &"
         sleep 1 # FIXME: qemu-nbd needs to be listening before we continue
     fi