Comment 31 for bug 1711203

Reviewing @slangasek's notes

> It's worth checking whether this problem
> mysteriously resolves once linux-signed is being pulled in; if it does,
> then it's possible we have a bug in grub (enforcing signature when it's
> not supposed to) or simply a bug in firmware.

It would appear that despite the change to linux-signed, there is still a
bug.
In that light, can we get next steps on debugging grub or firmware or
whateever
else is needed to push this along?

On Tue, Dec 5, 2017 at 7:58 AM, Rod Smith <email address hidden> wrote:

> I'd just like to emphasize that, although a change to always install the
> linux-signed kernel on AMD64 systems is necessary to fix this bug, it's
> not sufficient to fix the bug. As noted in my comment #25 (and
> elsewhere), another change is also required -- either a change to Shim
> or GRUB (I don't know which) or a change to how MAAS handles the boot
> process (to have the PXE-booted GRUB read the configuration file from
> the hard disk rather than chainload to GRUB on the hard disk; or perhaps
> a change to the way the handoff is done, if some tweak could bypass the
> bug).
>
> As before, I remain able and willing to test potential fixes.
>
> --
> You received this bug notification because you are subscribed to curtin.
> Matching subscriptions: curtin-bugs-all
> https://bugs.launchpad.net/bugs/1711203
>
> Title:
> Deployments fail when Secure Boot enabled
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/curtin/+bug/1711203/+subscriptions
>