[Libguestfs] [PATCH v2 3/3] v2v: -o rhv-upload: Add a test.

Nir Soffer nsoffer at redhat.com
Tue Oct 16 07:32:30 UTC 2018


On Mon, Oct 15, 2018 at 7:13 PM Richard W.M. Jones <rjones at redhat.com>
wrote:

> On Mon, Oct 15, 2018 at 06:43:10PM +0300, Nir Soffer wrote:
> > On Mon, Oct 15, 2018 at 6:21 PM Richard W.M. Jones <rjones at redhat.com>
> > wrote:
> >
> > > On Tue, Oct 09, 2018 at 02:28:10PM +0300, Nir Soffer wrote:
> > > > > +# Create a background thread running a web server which is
> > > > > +# simulating the imageio server.
> > > > >
> > > >
> > > > This functionality should be separated from the fake SDK module,
> since
> > > it is
> > > > not part of the SDK, and may be replaced by real imageio server
> later.
> > >
> > > Well possibly, but it was very convenient to put it in the class here,
> > > and this test is meant for running completely standalone without any
> > > other service available.
> > >
> > > > > +class RequestHandler(BaseHTTPRequestHandler):
> > > > >
> > > >
> > > > This request handler is using HTTP/1.0, and will close the connection
> > > after
> > > > every request. This is not a good implementation of the imageio
> server,
> > > and
> > > > also
> > > > hides bugs in this code.
> > > >
> > > > Should be fixed by adding:
> > > >
> > > >     protocol_version = "HTTP/1.1"
> > >
> > > I tried the attached patch, but for some reason it just hangs.
> > >
> >
> > The issue is probably missing "content-length: 0" header in the
> > response. If we don't close the connection and don't send
> > content-length the client cannot do much but wait :-)
>
> Ah yes, it was missing from the OPTIONS method.
> The attached patch does work.  What do you think?
>

Looks good

Nir
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/libguestfs/attachments/20181016/3afdc13f/attachment.htm>


More information about the Libguestfs mailing list