Bug 1403733

Summary: RHEL6.9 installed in UEFI mode can't create boot entry in NVRAM on Lenovo Purley platform
Product: Red Hat Enterprise Linux 6 Reporter: Neo Cui <cuilj2>
Component: efibootmgrAssignee: Peter Jones <pjones>
Status: CLOSED WONTFIX QA Contact: Release Test Team <release-test-team>
Severity: high Docs Contact:
Priority: high    
Version: 6.9CC: cuilj2, fchen16, hehy1, jkachuck, laokunlun.tech, mknutson, pjones, tumeya, yangyg3
Target Milestone: rc   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-04-05 21:08:59 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Bug Depends On:    
Bug Blocks: 1415214, 1438054    
Attachments:
Description Flags
sosreport file none

Description Neo Cui 2016-12-12 10:01:38 UTC
Description of problem:
When I fresh install RHEL6.9 in UEFI mode, there is no boot entry for RHEL in the NVRAM. So I can't boot into OS successfully.

Version-Release number of selected component (if applicable):
efibootmgr-0.5.4

How reproducible:


Steps to Reproduce:
1. Set BIOS to UEFI mode
2. Fresh install the RHEL6.9
3. Then I cannot boot into OS as there is no boot entry for RHEL6.9

Actual results:
I cannot boot into RHEL6.9 after installation.

Expected results:
Boot into OS normally.

Additional info:

Comment 1 Neo Cui 2016-12-12 10:08:08 UTC
I can boot the OS from BIOS - "Boot from file" or "Add the UEFI boot path manually".
When I boot into OS, I found that there is a boot entry variable Boot0003xxxx which is related to Network devices created by BIOS. It's size is 3.3k large than 1k size limitation.
So I think it's this variable affected the creation of RHEL6.9 boot entry.
But I can't figure out the root cause in the efibootmgr source code.
So I need help from Red Hat.

Comment 4 Joseph Kachuck 2016-12-13 17:41:24 UTC
Hello Lenovo,
Please boot the system in rescue mode. Create and attach a sosreport from this mode.
This this give us the anaconda logs and additional information that will assist in working this issue.

Thank You
Joe Kachuck

Comment 5 Joseph Kachuck 2016-12-13 17:43:36 UTC
Hello Lenovo,
Please also confirm is this is the same issue as:
https://bugzilla.redhat.com/show_bug.cgi?id=1401776

Thank You
Joe Kachuck

Comment 6 Neo Cui 2016-12-14 00:41:37 UTC
(In reply to Joseph Kachuck from comment #5)
> Hello Lenovo,
> Please also confirm is this is the same issue as:
> https://bugzilla.redhat.com/show_bug.cgi?id=1401776
> 
> Thank You
> Joe Kachuck

I don't think it is the same issue.
One is boot entry creation problem. The other one is ACPI issue.
I think you may paste a wrong bug number.

Comment 7 Joseph Kachuck 2016-12-14 21:16:13 UTC
Hello Lenovo,
Please boot the system in rescue mode. Create and attach a sosreport from this mode.
This this give us the anaconda logs and additional information that will assist in working this issue.

Thank You
Joe Kachuck

Comment 8 steel 2016-12-15 01:57:01 UTC
Created attachment 1231941 [details]
sosreport file

Comment 9 steel 2016-12-15 01:59:41 UTC
The sosreport md5 is 0db777cf181dcc455ae4eb2b38ca8f3b

Comment 10 Neo Cui 2016-12-20 05:48:49 UTC
Can you help us understand the detailed OS handling:
1. If Boot#### exist and less than 1K size, BIOS will return EFI_SUCCESS(0)
2. If Boot#### not exist, BIOS will return EFI_NOT_FOUND(0x8000000E)
3. If Boot#### exist and larger than 1K size, BIOS will return EFI_BUFFER_TOO_SMALL(0x80000005)

How do OS handle these 3 cases respectively?

Comment 11 Joseph Kachuck 2017-02-14 19:55:59 UTC
Hello,
Due to were are for the RHEL 6.9 release. This will not make RHEL 6.9. This is being moved to RHEL 6.10.

Thank You
Joe Kachuck

Comment 12 Neo Cui 2017-02-15 00:42:58 UTC
(In reply to Joseph Kachuck from comment #11)
> Hello,
> Due to were are for the RHEL 6.9 release. This will not make RHEL 6.9. This
> is being moved to RHEL 6.10.
> 
> Thank You
> Joe Kachuck

Hi Joseph,

Do you mean RHEL 6.10 will introduce the new EFI variable interface - efivarfs?

Comment 13 Joseph Kachuck 2017-02-15 15:15:22 UTC
Hello Lenovo,
Please confirm to correct this issue. It would be required to add a new EFI variable interface? This is something that was not included in RHEL 6.x previously?

At current this is requested for RHEL 6.10.

Thank You
Joe Kachuck

Comment 14 Neo Cui 2017-02-16 00:40:58 UTC
(In reply to Joseph Kachuck from comment #13)
> Hello Lenovo,
> Please confirm to correct this issue. It would be required to add a new EFI
> variable interface? This is something that was not included in RHEL 6.x
> previously?
> 
> At current this is requested for RHEL 6.10.
> 
> Thank You
> Joe Kachuck

Hi Joseph,

This defect is caused by 1K size limitation of EFI variable.
If there is a boot entry variable large than 1K, the defect will occur.
So the new EFI variable interface - efivarfs should be introduced in kernel.
It would better if it can be fixed in RHEL 6.9.
If there is no possibility to fix this defect in RHEL 6.9, please add the request in 6.10.
Thanks~

Comment 15 Kunlun Lao 2017-03-15 17:27:23 UTC
The 1KB variable problem had been fixed in previous version.

* Thu Sep 04 2014 Peter Jones <pjones> - 0.5.4-12
- Display UEFI boot variables when the contents are > 1024 bytes
  Resolves: rhbz#1121782

Comment 17 Joseph Kachuck 2017-04-05 21:08:59 UTC
Hello,
RHEL 6 has entered Phase 3. In phase 3 only Critical impact Security Advisories and selected Urgent Priority Bug Fix Advisories will be accepted.
https://access.redhat.com/support/policy/updates/errata

At current this BZ does not meet these requirements. I am closing this BZ as WONTFIX.

Please reopen if this fix is required for RHEL 6. If so please also provide a justification for this fix.

Thank You
Joe Kachuck