On 11/11/19 4:04 AM, Andrey Shinkevich wrote:
The test for an NBD client. The NBD server is disconnected after the
client write request. The NBD client should reconnect and complete
the write operation.

Suggested-by: Denis V. Lunev <[email protected]>
Suggested-by: Vladimir Sementsov-Ogievskiy <[email protected]>
Signed-off-by: Andrey Shinkevich <[email protected]>
---
v3 was discussed in the email thread with the message ID
<1572627272-23359-1-git-send-email-andrey.shinkev...@virtuozzo.com>


+import os
+import subprocess
+import iotests
+from iotests import file_path, log
+
+
+nbd_sock, conf_file = file_path('nbd-sock', "nbd-fault-injector.conf")

Mixing '' and "" in the same line is odd. I don't know if we have a strong preference for one style over the other.

+++ b/tests/qemu-iotests/group
@@ -284,3 +284,4 @@
 268 rw auto quick
 270 rw backing quick
 272 rw
+277 rw

The test completes in 3 seconds for me; we could add 'quick'.

Otherwise, looks reasonable to me.

Reviewed-by: Eric Blake <[email protected]>
Tested-by: Eric Blake <[email protected]>

I'll probably queue this through my NBD tree for rc1, as it adds test coverage of a new feature for the 4.2 release.

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


Reply via email to