kernel panic after booting up 4 hours

[Alfred]

After upgrading to Wily from Vivid, I always meet kernal panic after booting up 4 hours;

I use wireless, the Ethernet cable is not connected.

ProblemType: Bug
DistroRelease: Ubuntu 15.10
Package: linux-image-4.2.0-16-generic 4.2.0-16.19
ProcVersionSignature: Ubuntu 4.2.0-16.19-generic 4.2.3
Uname: Linux 4.2.0-16-generic x86_64
ApportVersion: 2.19.1-0ubuntu3
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: alfred 5470 F.... pulseaudio
 /dev/snd/controlC0: alfred 5470 F.... pulseaudio
CurrentDesktop: Unity
Date: Mon Oct 26 12:57:59 2015
HibernationDevice: RESUME=UUID=a9831098-f784-41f9-b1f4-7f39b1040838
InstallationDate: Installed on 2015-01-23 (275 days ago)
InstallationMedia: Ubuntu 14.10 "Utopic Unicorn" - Release amd64 (20141022.1)
MachineType: LENOVO 20ANS01N00
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.2.0-16-generic.efi.signed root=UUID=cf490836-5dbd-4c19-8796-52575400a1f9 ro nomdmonddf nomdmonisw
RelatedPackageVersions:
 linux-restricted-modules-4.2.0-16-generic N/A
 linux-backports-modules-4.2.0-16-generic N/A
 linux-firmware 1.149
SourcePackage: linux
UpgradeStatus: Upgraded to wily on 2015-10-23 (2 days ago)
dmi.bios.date: 06/23/2015
dmi.bios.vendor: LENOVO
dmi.bios.version: GLET79WW (2.33 )
dmi.board.asset.tag: Not Available
dmi.board.name: 20ANS01N00
dmi.board.vendor: LENOVO
dmi.board.version: SDK0A46860 WIN
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Not Available
dmi.modalias: dmi:bvnLENOVO:bvrGLET79WW(2.33):bd06/23/2015:svnLENOVO:pn20ANS01N00:pvrThinkPadT440p:rvnLENOVO:rn20ANS01N00:rvrSDK0A46860WIN:cvnLENOVO:ct10:cvrNotAvailable:
dmi.product.name: 20ANS01N00
dmi.product.version: ThinkPad T440p
dmi.sys.vendor: LENOVO

[Alfred (alf-redyoung)]

seem to be related(alike):
http://lkml.iu.edu/hypermail/linux/kernel/1508.2/01174.html
http://www.gossamer-threads.com/lists/linux/kernel/2198770

[Brad Figg (brad-figg)]

This change was made by a bot.

[Stefan Bader (smb)]

Checking the referred links it sounds indeed like the following patch is required for the e1000e driver (even or because it is not in use). Seems like it has been asked to be included in 4.2 before release but nothing happened. Did not see it either in Dave Millers most recent stable mbox. I am preparing a test kernel and will update this report with a link once done.

From 37b12910dd11d9ab969f2c310dc9160b7f3e3405 Mon Sep 17 00:00:00 2001
From: Raanan Avargil <email address hidden>
Date: Sun, 19 Jul 2015 16:33:20 +0300
Subject: [PATCH] e1000e: Fix tight loop implementation of systime read
 algorithm

Change the algorithm. Read systimel twice and check for overflow.
If there was no overflow, use the first value.
If there was an overflow, read systimeh again and use the second
systimel value.

[Stefan Bader (smb)]

@Alfred, could you try the packages at http://people.canonical.com/~smb/lp1509929/ and report back here whether this prevents the problem for you? Thanks,

[Alfred (alf-redyoung)]

Let me try; when I wake up, we can see if it is ok.

[Alfred (alf-redyoung)]

@Stefan, these packages look great. My laptop has worked over 4 hours.

$ uname -rv
4.2.0-17-generic #21+lp1509929v1 SMP Mon Oct 26 10:33:08 UTC 2015
$ uptime
 07:28:04 up 9:05, 2 users, load average: 0.34, 0.49, 0.26
$ lsmod | grep e1000e
e1000e 237568 0
ptp 20480 1 e1000e

[Stefan Bader (smb)]

Thanks for the update, I will submit the missing patch to upstream stable and in parallel try to get it included in the distro kernel as soon as possible.

[Stefan Bader (smb)]

SRU justification:

impact: A change in the e1000e network driver can cause a panics after 4 hours of uptime if the wired network is not in use.

Fix: Cherry-picking a follow-up patch from 4.3-rc1 avoids the issue.

Testcase: Leave a machine with e1000e hw not in use running for more than 4 hours.

[Stefan Bader (smb)]

I am setting this to fix-released as it is so in 4.3 and sooner than later the Xenial kernel will move to be 4.3 based and then have the fix. Unfortunately there is no automatic way of closing the bug report then and could be easily forgotten.

[Alfred (alf-redyoung)]

I've verified latest wily kernel, it fix this problem.

$ uname -rv
4.2.0-19-generic #23-Ubuntu SMP Wed Nov 11 11:39:30 UTC 2015
$ uptime
 07:43:26 up 11:42, 10 users, load average: 0.27, 0.11, 0.08
$ lsmod | grep e1000
e1000e 237568 0
ptp 20480 1 e1000e

[Olivier Louvignes (olouvignes)]

Looks like I'm encountering the exact same issue, in my case it's not every single day, but happen quite often ~10% of our devices daily. I opened a new bug report: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1668356

[Alfred (alf-redyoung)]

@Olivier, I never see such problem for a long time, after 4.2.0.19; I've upgraded to Xenial, current using kernel is 4.4.0.62, works fine; Kernel 4.4.0.64 is installed, but I do not reboot my laptop yet.