Message ID | 20180205125023.12394-1-vsementsov@virtuozzo.com |
---|---|
State | New |
Headers | show |
Series | iotests: 205: support only raw format | expand |
On 02/05/2018 06:50 AM, Vladimir Sementsov-Ogievskiy wrote: > The test is about qmp interface to NBD server, so there no reasons > to run it for various disk formats. Also it actually do not support > luks format, so, disable all format except raw. > > Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> > --- > tests/qemu-iotests/205 | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/tests/qemu-iotests/205 b/tests/qemu-iotests/205 > index 10388920dc..a78ff37104 100644 > --- a/tests/qemu-iotests/205 > +++ b/tests/qemu-iotests/205 > @@ -153,4 +153,4 @@ class TestNbdServerRemove(iotests.QMPTestCase): > > > if __name__ == '__main__': > - iotests.main() > + iotests.main(supported_fmts=['raw']) Isn't that stricter than necessary? The test works with qcow2; it was just luks that was special (because we weren't passing in proper secrets). At any rate, I'm adding Dan in case he has a quick idea of what it would take to get luks working, rather than just disabling it.
On Mon, Feb 05, 2018 at 09:40:20AM -0600, Eric Blake wrote: > On 02/05/2018 06:50 AM, Vladimir Sementsov-Ogievskiy wrote: > > The test is about qmp interface to NBD server, so there no reasons > > to run it for various disk formats. Also it actually do not support > > luks format, so, disable all format except raw. > > > > Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> > > --- > > tests/qemu-iotests/205 | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/tests/qemu-iotests/205 b/tests/qemu-iotests/205 > > index 10388920dc..a78ff37104 100644 > > --- a/tests/qemu-iotests/205 > > +++ b/tests/qemu-iotests/205 > > @@ -153,4 +153,4 @@ class TestNbdServerRemove(iotests.QMPTestCase): > > > > > > if __name__ == '__main__': > > - iotests.main() > > + iotests.main(supported_fmts=['raw']) > > Isn't that stricter than necessary? The test works with qcow2; it was > just luks that was special (because we weren't passing in proper > secrets). At any rate, I'm adding Dan in case he has a quick idea of > what it would take to get luks working, rather than just disabling it. For shell based tests it is not too difficult as I've got much fo the support work done. This is a python tests though, so would need some prep work first. The key is that you need to be able to add extra args to QEMU to define the secret with --object, and use the full blockdev options based syntax for -drive. I'm not expecting you todo that unless you're actively interested in it, so no objection to just disabling it I will point out though there is an 'unsupported_fmts' arg to the python tests, so a flexible fix would be to invert your patch todo unsupported_fmts=['luks'] Regards, Daniel
diff --git a/tests/qemu-iotests/205 b/tests/qemu-iotests/205 index 10388920dc..a78ff37104 100644 --- a/tests/qemu-iotests/205 +++ b/tests/qemu-iotests/205 @@ -153,4 +153,4 @@ class TestNbdServerRemove(iotests.QMPTestCase): if __name__ == '__main__': - iotests.main() + iotests.main(supported_fmts=['raw'])
The test is about qmp interface to NBD server, so there no reasons to run it for various disk formats. Also it actually do not support luks format, so, disable all format except raw. Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com> --- tests/qemu-iotests/205 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)