Bug 25080

Summary:	Update request: systemd-241-8.1.mga7
Product:	Mageia	Reporter:	Thomas Backlund <tmb>
Component:	RPM Packages	Assignee:	QA Team <qa-bugs>
Status:	RESOLVED FIXED	QA Contact:
Severity:	critical
Priority:	High	CC:	crxssi, davidwhodgins, jim, sysadmin-bugs
Version:	7	Keywords:	advisory, validated_update
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:	MGA7-64-OK
Source RPM:	systemd	CVE:
Status comment:
Bug Depends on:
Bug Blocks:	25027

Description Thomas Backlund 2019-07-09 21:29:49 CEST

Advisory:
Updated systemd packages fixes booting on new AMD Ryzen systems

On some AMD CPUs RDRAND will sometimes report success via the carry flag
but nonetheless return the same fixed value -1 in all cases. This appears
to be a bad bug in the CPU or firmware. This causes systems to fail to boot
after suspend/resume.

With the new Amd Ryzen 3xxx series that got released on July 7th, 2019 its
worse, as the system will fail to boot every time because no systemd service
will start properly.

This update adds a workaround to systemd so it ignores the invalid RDRAND
responses, thus allowing the system to boot.

In this update there is also a fix for udevd having a reversed conditional on
global property set, report causing some ENV variables being reported as not
set even if they are.


SRPMS:
systemd-241-8.1.mga7.src.rpm

i586:
libsystemd0-241-8.1.mga7.i586.rpm
libudev1-241-8.1.mga7.i586.rpm
libudev-devel-241-8.1.mga7.i586.rpm
nss-myhostname-241-8.1.mga7.i586.rpm
systemd-241-8.1.mga7.i586.rpm
systemd-devel-241-8.1.mga7.i586.rpm
systemd-tests-241-8.1.mga7.i586.rpm
systemd-units-241-8.1.mga7.i586.rpm

x86_64:
lib64systemd0-241-8.1.mga7.x86_64.rpm
lib64udev1-241-8.1.mga7.x86_64.rpm
lib64udev-devel-241-8.1.mga7.x86_64.rpm
nss-myhostname-241-8.1.mga7.x86_64.rpm
systemd-241-8.1.mga7.x86_64.rpm
systemd-devel-241-8.1.mga7.x86_64.rpm
systemd-tests-241-8.1.mga7.x86_64.rpm
systemd-units-241-8.1.mga7.x86_64.rpm

Thomas Backlund 2019-07-09 21:29:57 CEST

Priority: Normal => High

Comment 1 Thomas Backlund 2019-07-09 21:33:56 CEST

Note to testers, The Ryzen non-booting bug requires a AMD Ryzen 3xxx cpu to verify.

I have done so today with a brand new Ryzen 5 3600 that arrived here today.

So just install this update, and test that nothing breaks.

The udevd bug will hopefully be verified in bug 25027

Blocks: (none) => 25027

Comment 2 Thomas Backlund 2019-07-09 21:35:13 CEST

(In reply to Thomas Backlund from comment #1)
> 
> The udevd bug will hopefully be verified in bug 25027

Meaning I hope it will fix that bug.
Regardless, its the correct fix to do and comes from upstream

Comment 3 Thomas Backlund 2019-07-09 22:17:36 CEST

(In reply to Thomas Backlund from comment #2)
> (In reply to Thomas Backlund from comment #1)
> > 
> > The udevd bug will hopefully be verified in bug 25027
> 
> Meaning I hope it will fix that bug.

Martinw just confirmed the udevd bug is fixed with this update

Comment 4 Rémi Verschelde 2019-07-10 11:03:45 CEST

Advisory uploaded.

Keywords: (none) => advisory

Comment 5 Rémi Verschelde 2019-07-10 12:10:06 CEST

Installed on the following system, reboot went fine and everything seems to work.

$ inxi
CPU: Dual Core Intel Core i5-3230M (-MT MCP-) speed/min/max: 2278/1200/2600 MHz Kernel: 5.1.14-desktop-1.mga7 x86_64 Up: 15m 
Mem: 2113.6/11905.3 MiB (17.8%) Storage: 465.76 GiB (74.2% used) Procs: 224 Shell: bash 4.4.23 inxi: 3.0.33

Comment 6 Thomas Backlund 2019-07-10 12:14:07 CEST

Note that this update already is in Cauldron since Mon 08 Jul 2019 11:55:48 PM CEST, so it's tested there too

Comment 7 M D 2019-07-10 22:59:06 CEST

(In reply to Thomas Backlund from comment #1)
> Note to testers, The Ryzen non-booting bug requires a AMD Ryzen 3xxx cpu to
> verify.
> 
> I have done so today with a brand new Ryzen 5 3600 that arrived here today.

(In reply to Thomas Backlund from comment #6)
> Note that this update already is in Cauldron since Mon 08 Jul 2019 11:55:48
> PM CEST, so it's tested there too

Thomas, thanks so much for reporting the bug and the testing and followup.  I posted my advisory on Jul 8th on the forum: 

https://forums.mageia.org/en/viewtopic.php?f=8&t=12894

It is great it is already in Cauldron.  I assume your testing showed it not possible to boot/install/run Mageia 7.0.  Will Mageia have to release new 7.0 ISO images?  If not, how will Ryzen 3000 users work-around the bug to install 7.0?

PS- what motherboard did you use?  Is it X570 based?

CC: (none) => crxssi

Comment 8 James Kerr 2019-07-11 10:57:04 CEST

On mga7-64

packages installed cleanly:
- lib64systemd0-241-8.1.mga7.x86_64
- lib64udev1-241-8.1.mga7.x86_64
- nss-myhostname-241-8.1.mga7.x86_64
- systemd-241-8.1.mga7.x86_64
- systemd-units-241-8.1.mga7.x86_64

no regressions observed
looks OK for mga7-64 on this system:

Dell product: Precision Tower 3620
Mobo: Dell model: 09WH54 
Card: Intel HD Graphics 530
CPU: Quad core Intel Core i7-6700 (-HT-MCP-)
PC-BIOS (legacy) boot
GPT partitions

(I've also been running systemd-241-9 on my cauldron install on this machine  for two days)

CC: (none) => jim

Dave Hodgins 2019-07-11 21:10:19 CEST

Whiteboard: (none) => MGA6-64-OK MGA7-64-OK
Keywords: (none) => validated_update
CC: (none) => davidwhodgins, sysadmin-bugs

Comment 9 Mageia Robot 2019-07-11 22:51:20 CEST

An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGAA-2019-0062.html

Status: NEW => RESOLVED
Resolution: (none) => FIXED

Rémi Verschelde 2019-07-12 11:19:30 CEST

Whiteboard: MGA6-64-OK MGA7-64-OK => MGA7-64-OK