[PATCH] Subdomains: Send the DP reply in the correct format

[Jakub Hrozek]

I found this issue when testing Simo's patches.

The DP was sending the reply in a format the responder did not expect,
so the responder always failed to parse the message.

[Pavel Březina]

On 07/31/2012 01:28 PM, Jakub Hrozek wrote:
> I found this issue when testing Simo's patches.
>
> The DP was sending the reply in a format the responder did not expect,
> so the responder always failed to parse the message.

Nack.

You should talloc_free(req) even when you are unable to send dbus 
message, because it is the end of the request anyway.

[Jakub Hrozek]

On Thu, Aug 02, 2012 at 11:00:58AM +0200, Pavel Březina wrote:
> On 07/31/2012 01:28 PM, Jakub Hrozek wrote:
> > I found this issue when testing Simo's patches.
> >
> > The DP was sending the reply in a format the responder did not expect,
> > so the responder always failed to parse the message.
>
> Nack.
>
> You should talloc_free(req) even when you are unable to send dbus
> message, because it is the end of the request anyway.

Thanks, new patch is attached.

[Pavel Březina]

On 08/07/2012 09:13 AM, Jakub Hrozek wrote:
> On Thu, Aug 02, 2012 at 11:00:58AM +0200, Pavel Březina wrote:
> > On 07/31/2012 01:28 PM, Jakub Hrozek wrote:
> > > I found this issue when testing Simo's patches.
> > >
> > > The DP was sending the reply in a format the responder did not expect,
> > > so the responder always failed to parse the message.
> >
> > Nack.
> >
> > You should talloc_free(req) even when you are unable to send dbus
> > message, because it is the end of the request anyway.
>
> Thanks, new patch is attached.

Ack.

[Jakub Hrozek]

On Tue, Aug 07, 2012 at 09:53:05AM +0200, Pavel Březina wrote:
> On 08/07/2012 09:13 AM, Jakub Hrozek wrote:
> > On Thu, Aug 02, 2012 at 11:00:58AM +0200, Pavel Březina wrote:
> > > On 07/31/2012 01:28 PM, Jakub Hrozek wrote:
> > > > I found this issue when testing Simo's patches.
> > > >
> > > > The DP was sending the reply in a format the responder did not expect,
> > > > so the responder always failed to parse the message.
> > >
> > > Nack.
> > >
> > > You should talloc_free(req) even when you are unable to send dbus
> > > message, because it is the end of the request anyway.
> >
> > Thanks, new patch is attached.
>
> Ack.

Pushed to master.