During the installation under UEFI mode, "Could not open \EFI\BOOT\fallback.efi: 14" shows after passed the BIOS phase.

Please also refer to the attached screenshot for the issue.

The attachment "no-output-on-fallback-checking.patch" seems to be a patch. If it isn't, please remove the "patch" flag from the attachment, remove the "patch" tag, and if you are a member of the ~ubuntu-reviewers, unsubscribe the team.

[This is an automated message performed by a Launchpad user owned by ~brian-murray, for any issues please contact him.]

Note that this is already fixed in bzr (lp:ubuntu/shim) by building fallback.efi; but this will require a round-trip to Microsoft for signing to get this updated in our images.

Hi Steve Langasek, thank you, just want to double confirm one thing, I checked lp:ubuntu/shim, the last commit is:

79. By Steve Langasek on 2013-09-27
Install fallback.efi.signed as well, to lay the groundwork for fallback
handling (wanted when we have to move a drive between machines, or when

And we still can find those Print() in the source code.

Could you give me more hints how is this issue get fixed? Thanks.

Yu,

The point is that the message in question is shown because of a legitimate misconfiguration: fallback.efi should be present and is not. Once that is corrected, there is no need to remove the Print() call.

The other Print() calls removed in your patch are not shown in your screenshot, so I'm not sure that they need to be removed either.

Steve, thanks for the explanation, now I'm clear of the solution, there is no need to remove any of the Print() calls.

From what I've interpreted reading about the purpose of fallback.efi, you genuinely do not want fallback.efi to exist when you are booting from removable media. Only when booting from the existing disk that you may or may not have moved to another machine.

The portion of the submitted patch that removed the error message is exactly the same thing that ended up upstream.

https://github.com/mjg59/shim/commit/41e18d4334d84cb5f7c7dd044c9afaf24a0a6538

This bug was fixed in the package shim - 0.7-0ubuntu4

---------------
shim (0.7-0ubuntu4) utopic; urgency=medium

  * SECURITY UPDATE: heap overflow and out-of-bounds read access when
    parsing DHCPv6 information
    - debian/patches/CVE-2014-3675.patch: apply proper bounds checking
      when parsing data provided in DHCPv6 packets.
    - CVE-2014-3675
    - CVE-2014-3676
  * SECURITY UPDATE: memory corruption when processing user-provided key
    lists
    - debian/patches/CVE-2014-3677.patch: detect malformed machine owner
      key (MOK) lists and ignore them, avoiding possible memory corruption.
    - CVE-2014-3677

shim (0.7-0ubuntu2) utopic; urgency=medium

  * Restore debian/patches/prototypes, which still is needed on shim 0.7
    but only detected on the buildds.
  * Update debian/patches/prototypes with some new declarations needed for
    openssl 0.9.8za update.

shim (0.7-0ubuntu1) utopic; urgency=medium

  * New upstream release.
    - fix spurious error message when fallback.efi is not present, as will
      always be the case for removable media. LP: #1297069.
    - drop most patches, included upstream.
  * debian/patches/0001-Update-openssl-to-0.9.8za.patch: cherry-pick
    openssl 0.9.8za in via upstream.

shim (0.4-0ubuntu5) utopic; urgency=low

  * Install fallback.efi.signed as well, to lay the groundwork for fallback
    handling (wanted when we have to move a drive between machines, or when
    the firmware loses its marbles^W nvram).
 -- Steve Langasek <email address hidden> Wed, 08 Oct 2014 06:40:40 +0000