diff --git a/src/xenvif/frontend.c b/src/xenvif/frontend.c
index 5a29736..aef6680 100644
--- a/src/xenvif/frontend.c
+++ b/src/xenvif/frontend.c
@@ -1521,6 +1521,11 @@ FrontendPrepare(
FrontendSetXenbusState(Frontend,
XenbusStateInitialising);
break;
+
+ case XenbusStateConnected:
+ FrontendSetXenbusState(Frontend,
+ XenbusStateClosing);
+ break;
case XenbusStateClosing:
FrontendSetXenbusState(Frontend,
On Mon, Jul 18, 2016 at 11:44 AM, Paul Durrant <Paul.Durrant@xxxxxxxxxx> wrote:
>
> Sorry, I realised I had not responded to this. This does indeed look like a hole in the logic in the master branch. I will add your patch but since you did not add your S-o-b to the patch I can only credit you with a ‘suggested-by’. If you can post a patch
in the correct form (see
http://wiki.xenproject.org/wiki/Submitting_Xen_Project_Patches for guidance) then I can give you full credit.
>
>
>
> Cheers,
>
>
>
> Paul
>
>
>
> From: win-pv-devel [mailto:win-pv-devel-bounces@xxxxxxxxxxxxxxxxxxxx] On Behalf Of Eytan Heidingsfeld
> Sent: 11 July 2016 15:06
> To: win-pv-devel@xxxxxxxxxxxxxxxxxxxx
> Subject: [win-pv-devel] Patch for Unexpected Initial States of XenVif
>
>
>
> Hello,
> I've noticed that the support for unexpected initial states that was added to support iscsi boot in commit: c9b6e56e7342a1bb31cfb79f28b6a00d05954006,
>
>
>
> As there was quite a bit of refactoring around that area I've created a small patch that fixes the issue again,
>
>
>
> diff --git a/src/xenvif/frontend.c b/src/xenvif/frontend.c
>
> index 5a29736..aef6680 100644
>
> --- a/src/xenvif/frontend.c
>
> +++ b/src/xenvif/frontend.c
>
> @@ -1521,6 +1521,11 @@ FrontendPrepare(
>
> FrontendSetXenbusState(Frontend,
>
> XenbusStateInitialising);
>
> break;
>
> +
>
> + case XenbusStateConnected:
>
> + FrontendSetXenbusState(Frontend,
>
> + XenbusStateClosing);
>
> + break;
>
>
>
> case XenbusStateClosing:
>
> FrontendSetXenbusState(Frontend,
>
> --
>
> 1.9.5.msysgit.0
>
>
>
> Cheers,
>
> Eytan