System Hangs and General Protection Fault Occurs following NFS Access

[Mike Vastola]

Last night, in an attempt to secure my system against Meltdown/Spectre, I upgraded from Zesty to Artful. Upon booting into the latest kernel, my system would immediately hang and fail to respond (including to SysRq commands).

Both this (my primary) system and an Intel NUC (which were both upgraded from Zesty to Artful at the same time) export NFS shares to one other. The NUC appears to be error-free, but my primary system seems to crash whenever I attempt to access the NFS share of NUC.

As I have not tried any other kernel releases in either system other than 4.13.0-25.29, it's impossible for me to say if this is an issue regarding the recent security patch, or if it is some other issue.

Attached is the console log (obtained via netconsole) which includes the kernel trace of the error.

Please let me know if further information is required or if I can help further in troubleshooting.

ProblemType: Bug
DistroRelease: Ubuntu 17.10
Package: linux-image-4.13.0-25-generic 4.13.0-25.29
ProcVersionSignature: Ubuntu 4.10.0-42.46-generic 4.10.17
Uname: Linux 4.10.0-42-generic x86_64
NonfreeKernelModules: nvidia_uvm nvidia_drm nvidia_modeset nvidia
ApportVersion: 2.20.7-0ubuntu3.7
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC2: mvastola 4661 F.... pulseaudio
 /dev/snd/controlC1: mvastola 4661 F.... pulseaudio
CurrentDesktop: KDE
Date: Wed Jan 10 18:18:52 2018
InstallationDate: Installed on 2012-01-06 (2196 days ago)
InstallationMedia: Kubuntu 11.10 "Oneiric Ocelot" - Release amd64 (20111011)
IwConfig:
 lo no wireless extensions.

 mike no wireless extensions.

 docker0 no wireless extensions.
MachineType: Gigabyte Technology Co., Ltd. GA-990XA-UD3
ProcFB: 0 VESA VGA
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-4.10.0-42-generic root=/dev/mapper/VastDesk-RootOS ro apparmor=0 netconsole=4444@10.12.1.55/mike,4444@10.12.1.1/8c:ae:4c:f4:83:1a crashkernel=384M-:128M crashkernel=384M-:128M
RelatedPackageVersions:
 linux-restricted-modules-4.10.0-42-generic N/A
 linux-backports-modules-4.10.0-42-generic N/A
 linux-firmware 1.169.1
RfKill:
 0: hci0: Bluetooth
  Soft blocked: no
  Hard blocked: no
SourcePackage: linux
UpgradeStatus: Upgraded to artful on 2014-12-03 (1134 days ago)
dmi.bios.date: 09/09/2014
dmi.bios.vendor: Award Software International, Inc.
dmi.bios.version: F14e
dmi.board.name: GA-990XA-UD3
dmi.board.vendor: Gigabyte Technology Co., Ltd.
dmi.chassis.type: 3
dmi.chassis.vendor: Gigabyte Technology Co., Ltd.
dmi.modalias: dmi:bvnAwardSoftwareInternational,Inc.:bvrF14e:bd09/09/2014:svnGigabyteTechnologyCo.,Ltd.:pnGA-990XA-UD3:pvr:rvnGigabyteTechnologyCo.,Ltd.:rnGA-990XA-UD3:rvr:cvnGigabyteTechnologyCo.,Ltd.:ct3:cvr:
dmi.product.name: GA-990XA-UD3
dmi.sys.vendor: Gigabyte Technology Co., Ltd.

[Mike Vastola (michael-vastola)]

Please note that I ran 'ubuntu-bug' from my most recent working kernel (from Zesty): 4.10.0-42-generic, so some of this log output may be inaccurate as a result.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This change was made by a bot.

[Joseph Salisbury (jsalisbury)]

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.15 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.15-rc7

[Mike Vastola (michael-vastola)]

Okay, I tested it with 4.15.0-041500rc7-generic and the issue disappeared. Thanks!

Does that provide any insight into what exactly the error is and/or a possible workaround?

Unfortunately I'm unable to use this kernel in the interim as I rely on the Nvidia kernel modules, which aren't compiling in DKMS.

[Tetsuo Handa (9-launchpad-i-love-sakura-ne-jp)]

The dump looks like below bug. Was the bug already fixed in 4.13.0-25.29 ?

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1731031
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1734327

[Tetsuo Handa (9-launchpad-i-love-sakura-ne-jp)]

Forgot to mention. Try removing apparmor=0 parameter for now.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1734686

[Joseph Salisbury (jsalisbury)]

A way to see if this is related to the security patches is to test 4.13.0-21, which does not have the patches. It can be downloaded from:

https://launchpad.net/~canonical-kernel-team/+archive/ubuntu/ppa/+build/13870500

To test that kernel, install both the linux-image and linux-image-extra .deb packages.

If the bug still exists with -21, we can perform a "Reverse" kernel bisect to identify the commit that fixes this bug in mainline.

[Mike Vastola (michael-vastola)]

Okay, the bug still exists in 4.13.0-21-generic, but went away in 4.13.0-25-generic once I removed 'apparmor=0'. So I think Tetsuo may be on to something.

Could this be a duplicate of this[0] bug? This would make sense considering the message of one the problematic commits is "add stacking support to apparmor network hooks". I'm not well versed enough in the kernel internals to say for sure, so I'm deferring to someone who knows better.

Also, I'm not sure if I'm missing something, but while that bug is marked as "Fix Committed" in Launchpad[0], and a request was made to revert the pair of commits[1], I neither see a reply to that email saying the commits were indeed reverted nor subsequent commits to the Artful repo reverting the files affected by these commits. Is it possible that email got lost?

[0] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1734327
[1] https://lists.ubuntu.com/archives/kernel-team/2017-December/088715.html
[2] http://kernel.ubuntu.com/git/ubuntu/ubuntu-artful.git/log/security/apparmor/lsm.c and http://kernel.ubuntu.com/git/ubuntu/ubuntu-artful.git/log/security/selinux/netlabel.c

[Tetsuo Handa (9-launchpad-i-love-sakura-ne-jp)]

OK. Your report will be duplicate of 1731031, 1734327, 1734686.

Explanation of the bug is http://kernsec.org/pipermail/linux-security-module-archive/2017-November/004532.html
and the fix for the bug is http://kernsec.org/pipermail/linux-security-module-archive/2017-December/004638.html .

According to http://kernel.ubuntu.com/git/ubuntu/ubuntu-artful.git/log/security ,
neither the problematic patches are reverted nor the fix patch is applied.

[Mike Vastola (michael-vastola)]

Thanks!

Should I mark this bug as a duplicate then?

Also, what is the process/timeline for that patch you linked being applied downstream into Artful?

[Tetsuo Handa (9-launchpad-i-love-sakura-ne-jp)]

The problematic patches are not yet in mainline/upstream kernels,
and hence the fix patch is not yet in mainline/upstream kernels.

Please ask Joseph Salisbury when the action for Artful kernel will be taken.

[Mike Vastola (michael-vastola)]

Out of curiosity, can you link me to the upstream git repo for http://kernsec.org/pipermail/linux-security-module-archive/2017-December/004638.html? I want to be sure I'm looking at the right place.

Interestingly, it looks like someone replied to your patch saying they would incorporate it in early December, so it seems this bug fell through the cracks more than once.

Joseph Salisbury is subscribed to this bug and has commented on it above, so hopefully he will update regarding this when he gets a chance.

This seems like a really simple fix, so I'm actually going to see if I can recompile 4.13.0-25-generic with this patch added to make it work.

[Joseph Salisbury (jsalisbury)]

A SRU request has been submitted to have the problematic commit reverted until the permanent fix comes down from upstream.

I'll mark this bug as a duplicate of bug 1734327