segfault on start of live session in kactivitymanager / libqt5sql

Bug #1509839 reported by private_lock
16
This bug affects 2 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Expired
High
Unassigned

Bug Description

I've installed 15.10 Wily Werewolf to my USB stick via usb-creator and restart the laptop
on boot, I'm asked for the language and if I want to install or try a live session
I select try life session, the dialog disappears and the machine is frozen

Next try:
boot to selection dialog
switch to console crtl+alt+F1 (takes a moment)
issue alt+syreq+1 and alt+sysreq+t ... both answer "operation is disabled"
switch back to X ctrl+alt+F7 & click try live session
immediately switch back to console ctrl+alt+F1
watchdog detects hard lockup a few seconds later
see attached screenshot

reproduced also in beta1 of wily werewolf ... a month ago I imagine the beta was still starting, as for bug #1492840. But I've reformated the stick meanwhile and cannot really guarantee, that I didn't confuse the multiple ISO's I tried in short time.

version information:
kernel 4.2.0-16-generic #19-Ubuntu (4.1.0-3-generic #3-Ubuntu)
libqt5sql.so 5.4.2 (both)
currently running Kubuntu 15.04 on the same hardware
---
ApportVersion: 2.19.1-0ubuntu3
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC1: kubuntu 3331 F.... pulseaudio
 /dev/snd/controlC0: kubuntu 3331 F.... pulseaudio
CasperVersion: 1.365
DistroRelease: Ubuntu 15.10
LiveMediaBuild: Kubuntu 15.10 "Wily Werewolf" - Release amd64 (20151021)
MachineType: Micro-Star International Co., Ltd. GE70 2OC\2OD\2OE
Package: linux (not installed)
ProcEnviron:
 TERM=linux
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=de_DE.UTF-8
 SHELL=/bin/bash
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/casper/vmlinuz.efi file=/cdrom/preseed/username.seed boot=casper maybe-ubiquity quiet splash ---
ProcVersionSignature: Ubuntu 4.2.0-16.19-generic 4.2.3
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
Tags: wily
UdevLog: Error: [Errno 2] No such file or directory: '/var/log/udev'
Uname: Linux 4.2.0-16-generic x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo
_MarkForUpload: True
dmi.bios.date: 11/06/2014
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: E1757IMS.520
dmi.board.asset.tag: To be filled by O.E.M.
dmi.board.name: MS-1757
dmi.board.vendor: Micro-Star International Co., Ltd.
dmi.board.version: REV:0.B
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: Micro-Star International Co., Ltd.
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrE1757IMS.520:bd11/06/2014:svnMicro-StarInternationalCo.,Ltd.:pnGE702OC\2OD\2OE:pvrREV1.0:rvnMicro-StarInternationalCo.,Ltd.:rnMS-1757:rvrREV0.B:cvnMicro-StarInternationalCo.,Ltd.:ct10:cvrN/A:
dmi.product.name: GE70 2OC\2OD\2OE
dmi.product.version: REV:1.0
dmi.sys.vendor: Micro-Star International Co., Ltd.

Revision history for this message
private_lock (private-lock) wrote :
Revision history for this message
private_lock (private-lock) wrote :
Revision history for this message
private_lock (private-lock) wrote :
Revision history for this message
Brad Figg (brad-figg) wrote : Missing required logs.

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1509839

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

Changed in linux (Ubuntu):
status: New → Incomplete
Revision history for this message
Stefan Bader (smb) wrote :

Looks like the nouveau driver is involved in this somehow. Having a screenshot likely means at some point (when the cpu lockup is detected) there is an automatic switch to vt#1. Is it at that point possible to use alt+f2 (for example) to get to another text login screen and still login there (life iso user ubuntu and just enter for password). The other thing to try might be to interrupt the isolinux boot when the icons on the lower part of the screen appear with esc and then use nomodeset from the f6 menu. That only gives very low resolution but maybe gets to the live session.

Revision history for this message
private_lock (private-lock) wrote :

:~$ apport-collect 1509839
ERROR: Could not import module, is a package upgrade in progress? Error: No module named PyQt5.QtCore

I suspect, this was not, what you wanted to see? Oh ... I'm going to try to run it from the lifesession... but not this evening anymore, sorry.

@Stefan

Nope, there was no automatic switch. I did this myself manually right after clicking the offending "try Kubuntu" Button. Actually I prepared this step by switching back and forth to the console before, just to make sure, it will go smooth and not hang, while bash is still starting.

Gonna try Alt-F2 and nomodeset ... tomorrow

Revision history for this message
private_lock (private-lock) wrote : AlsaInfo.txt

apport information

tags: added: apport-collected wily
description: updated
Revision history for this message
private_lock (private-lock) wrote : CRDA.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : CurrentDmesg.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : IwConfig.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : JournalErrors.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : Lspci.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : Lsusb.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : ProcCpuinfo.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : ProcInterrupts.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : ProcModules.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : PulseList.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : RfKill.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : UdevDb.txt

apport information

Revision history for this message
private_lock (private-lock) wrote : WifiSyslog.txt

apport information

Revision history for this message
private_lock (private-lock) wrote :

1. Wow ... I got this apport-collect thingy running ... it won't run from any Konsole inside KDE, only the Ctrl+Alt+F1 text console will do. But phew ... it's asking to open a browser for login to ubuntu.com. Well I had to take another photo of the lengthy URL, switch back to X, launch firefox, type it in and allow the live session access to my account ... really intuitive!

2.Once I was playing around with switching between the consoles back and forth and got a soft-lockup only by trying to log in (sorry, no picture of that one). So while the message repeated every 20 seconds or so, I could still switch back and forth via Alt+F1, Alt+F2, Alt+F3, though at some point Alt+F2 stopped working and only Alt+F1, Alt+F3-6 showed distinct consoles. Trying to go to X via Alt+F7 I was stuck in a blank black screen and could not get back to text console.

3. (see screenshot): Once I got a softlockup, while bash was trying to figure out, that apport-collect is not installed by default. Actually the network cable was plugged out at that boot up. I didn't click on anything in the language chooser dialog, just waited for it to come up and switched directly to the console via Ctrl+Alt+F1 and logged in.

4. Right now I'm writing from inside a fully functional live-session. I didn't even need to experiment with nomodeset. The requirement is a true cold boot with about 30 seconds switched off completely. I did this 3 times in a row and it's rock solid. Whereas a cold boot after only a few seconds (really switching off and pressing the power button to start again) or a reboot via Alt+SyReq+b fails reliably (also 3 times in a row). It will go to the select language (English or German doesn't make a difference), then pressing try kubuntu, the screen goes all black (no mouse, no cursor, nothing) and it sits there forever. In addition, it won't accept keyboard commands like Ctrl+Alt+F1 to switch to the console or the good old Ctrl+Alt+Del. The only key combination, I could make work is Alt+SyReq+b, which obviously does not allow to investigate the error any further.

So as this is a fresh livesession, that I ran apport-collect, I doubt, there is any error in the logs, because my harddrive is encrypted and the USB stick is mounted readonly.

Oh and finally: For the last days, all theses experiments were done with the very same USB-stick. I haven't written anything to it and the md5-checksums are all OK. It's from this iso-image:
http://cdimage.ubuntu.com/kubuntu/releases/15.10/release/kubuntu-15.10-desktop-amd64.iso

Changed in linux (Ubuntu):
status: Incomplete → Confirmed
Revision history for this message
private_lock (private-lock) wrote :

OK, it not as "rock solid" ... right after writing the last comment, I did a restart and immediately came back into the live session. Thereafter it would not succeed again ... I did 6 cold reboots and each after a longer time period: 10 sec, 20 sec, 30 sec, 45 sec, 1 min, 5 min ... none would succeed.

Today, while the machine was still cold (low temperature), I did 5 reboots, by selecting reboot from inside the life session and I could not make it fail. The only difference I noticed is while waiting for the language selector, there is some variation: Either the screen is black or it already shows the background, sometimes it features a mouse cursor, that moves and sometimes not.

Stefan Bader (smb)
Changed in linux (Ubuntu):
assignee: nobody → Colin Ian King (colin-king)
Revision history for this message
Colin Ian King (colin-king) wrote :

The crash in comment #21 shows it is triggered by a read by thermald. To factor this out of the issue, can you temporarily disable thermald using:

sudo systemctl disable thermald

I doubt this is the root cause, but at least it factors out the crash you are seeing in comment #21 until we can figure out what else is going wrong.

Changed in linux (Ubuntu):
status: Confirmed → Incomplete
Revision history for this message
Colin Ian King (colin-king) wrote :

@private-lock, did you try out the instructions in comment #23? If so, did it help or not?

Revision history for this message
private_lock (private-lock) wrote :

Sorry for the long delay. Meanwhile I've upgraded the installed system to wily. It still has a few issues with network connections, but nothing like this bug.

Please find attached the screenshot after disabling thermald. The process is as usual:
- Boot from stick (no changes to kernel options) & wait for language selection
- ctrl-alt-F1 to console & type given command
- ctrl-alt-F7 to X & click "try kubuntu" & immediately switch back ctrl-alt-F1
- wait half a minute and see soft lockups
- enable trace by alt-sysreq-t
- take the picture

As you suspected a thermal problem, I took the notebook outside in the cold winter air: I've booted 15 to 20 times and could not produce a single lockup.

Changed in linux (Ubuntu):
status: Incomplete → New
Revision history for this message
Brad Figg (brad-figg) wrote : Status changed to Confirmed

This change was made by a bot.

Changed in linux (Ubuntu):
status: New → Confirmed
Changed in linux (Ubuntu):
importance: Undecided → Critical
Revision history for this message
penalvch (penalvch) wrote :

private_lock, could you please provide the full computer model as noted on the sticker of the computer itself (not from the Bug Description)?

Changed in linux (Ubuntu):
importance: Critical → High
status: Confirmed → Incomplete
Revision history for this message
private_lock (private-lock) wrote :

Sure!

vendor: msi
model: GE70 2oe
model-no 1757
CPU: Intel core i7 4700MQ
GPU: Nvidia GeForce GTX 765M / 2GB GDDR5
display: 17.3" FullHD non-glare
RAM: extended to 2x8 GB=16GB
HDD: 500GB
SSD: 250GB

More details and pictures at: http://de.msi.com/support/nb/GE70-2OE.html#down-bios

There is also a bunch of serial numbers hidden under the battery:
GE70 i765M245FD
PSN:001757-SKU8
4 719072 307363
S/N: 001757-SKU8K1401000043

Changed in linux (Ubuntu):
status: Incomplete → Confirmed
Changed in linux (Ubuntu):
assignee: Colin Ian King (colin-king) → nobody
Revision history for this message
penalvch (penalvch) wrote :

private_lock, your release has been EOL since July 28, 2016 as per https://wiki.ubuntu.com/Releases . Could you please advise if this is reproducible in a supported release?

tags: added: latest-bios-520
removed: hard kactivitymanager libqt5sql lockup segfault
Changed in linux (Ubuntu):
status: Confirmed → Incomplete
Revision history for this message
private_lock (private-lock) wrote :

I still own the same notebook so I could test it on the same hardware. But I wasn't able to freeze it a single time using multiple Kubuntu AMD64 ISO installed to USB-pendrives.

The following worked like a charm:
16.10
17.4
17.10
17.10.1
18.4.beta1

So I think you can safely close this bug.

Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

Changed in linux (Ubuntu):
status: Incomplete → Expired
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.