c9s: Secure Boot tests failing

[travier]

See: #1238

Needs investigation and opening a Bugzilla for the kernel in C9S.

We currently have to hardcode skipping the Secure Boot tests for C9S until we have coreos/coreos-assembler#3418.

[dustymabe]

Once we figure out the secureboot problem with c9s we should revert f779213 and db54842

[aaradhak]

The tests iso-live-login.uefi-secure & iso-as-disk.uefi-secure fail with c9s

[coreos-assembler]$ kola testiso -S iso-live-login.uefi-secure iso-as-disk.uefi-secure --output-dir tmp/kola-metal
⚠️  Skipping kola test pattern "ext.config.rpm-ostree.replace-rt-kernel":
  👉 https://github.com/openshift/os/issues/1099
Ignoring verification of signature on metal image
Running test: iso-as-disk.uefi-secure
FAIL: iso-as-disk.uefi-secure (10m0.002s)
    timed out after 10m0s
Running test: iso-live-login.uefi-secure
FAIL: iso-live-login.uefi-secure (10m0.002s)
    timed out after 10m0s
Error: harness: test suite failed
2023-06-09T17:21:53Z cli: harness: test suite failed

Console log output:

^[[2J^[[01;01H^[[=3h^[[2J^[[01;01H^[[2J^[[01;01H^[[=3h^[[2J^[[01;01HBdsDxe: loading Boot0001 "UEFI Misc Device" from PciRoot(0x0)/Pci(0x3,0x0)
BdsDxe: failed to load Boot0001 "UEFI Misc Device" from PciRoot(0x0)/Pci(0x3,0x0): Access Denied
BdsDxe: No bootable option or device was found.
BdsDxe: Press any key to enter the Boot Manager Menu.

[LorbusChris]

Not sure if related, but SCOS still pulls this ancient version of shim from the C9S repos:

Earliest changed package: shim-x64-15-15.el8_2.x86_64 at 2020-08-01 01:55:42 UTC

[dustymabe]

There may be something to this. I built with shim-x64-15.6-1.el9.x86_64.rpm from internal RHEL repos and it gets past the Access Denied. I wonder if the dbx in qemu in F38 (current COSA is based on F38) denies the older stuff from CentOS (similar to coreos/fedora-coreos-tracker#1452)

[dustymabe]

I opened this bug: https://bugzilla.redhat.com/show_bug.cgi?id=2213901

Let's see if that leads us anywhere.

[travier]

The old shim from RHEL 8 is "expected": https://bugzilla.redhat.com/show_bug.cgi?id=2115815

[dustymabe]

Talked to bstinson, he said they would take care of it. We'll wait on that before enabling signing for c9s.

[openshift-bot]

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle stale

[JanGutter]

I'm not 100% sure it's related, but I found out that I can't do secureboot in VM's in CentOS 9 Stream with the current edk2-ovmf firmware. There's a workaround available by installing the rpm from Rocky Linux, but the problem could definitely be elsewhere. I filed a bug on https://issues.redhat.com/browse/RHEL-11768

[travier]

/remove-lifecycle stale
/lifecycle frozen

[LorbusChris]

Tracked in https://issues.redhat.com/browse/RHEL-4391