apport information

** Attachment added: "ProcInterrupts.txt"
   
https://bugs.launchpad.net/bugs/1871766/+attachment/5350430/+files/ProcInterrupts.txt

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1871766

Title:
  Totem : Playing a corrupt video can result in full system crash -
  Linux Kernel unresponsive

Status in linux package in Ubuntu:
  Incomplete

Bug description:
  # DESCRIPTION :
  Playing a "partially corrupt" video starts normally, then starts showing 
visual artifacts (apparition of a red gradient on the bottom of the viewframe 
and a green gradient on top), which increase in intensity over time (can result 
in brigth blinking, black bars appearing in the middle of the video, etc.) and 
can, at some point, result in a complete system freeze.

  Once the system is frozen, the mouse can still move, but nothing else
  on the screen reacts - neither on click nor on hover. After a few
  seconds, the screens (I have two screens) go black briefly, then re-
  appear to show complete noise, with a 64x64-pixel-box of lighter noise
  around where the mouse is. The mouse can still move normally (the
  lighter 64x64 box moves at the same speed my mouse moved before the
  crash - no lag, no FPS drop).

  
  # IMPORTANT BEHAVIORAL OBSERVATIONS :
  - Alt + SysReq + R, E, I, S, U, B DOES NOT RESPOND (that's why I believe the 
kernel is affected). I tried several times, waiting a full second between each 
input; no response. I believe I was doing it correctly because I have used that 
technique successfully several times in the past on different systems.
  - The bug first occurred a few months ago, and since then I have updated my 
system several times (using Ubuntu's built-in graphical update notifier - I've 
installed everything update I saw), and none of the updates have fixed the 
problem. I did notice, though, that lately the bug was harder to reproduce, but 
I have no idea if that correlation means causation or not.
  - The same physical machine runs another instance of the exact same operating 
system (Ubuntu 18.04.4 LTS) on a different hard drive. I just switch hard 
drives manually to change systems. The other system has always used VLC to play 
videos as Totem lacks codecs to play the videos I want. It (the other system) 
suffers no major glitch, AFAIK.

  
  # STEPS TO REPRODUCE (might not work on different system - read further for 
why I believe it is necessary to report nevertheless):
  - Play a partially corrupt video file (I provided two in the attachment) with 
Totem and wait. I believe - but I am not sure - that stressing the application 
(for example, dragging the video scroller while the video plays) increases the 
likeliness of visual artifacts appearing. Usually, the green-at-the-top and 
red-at-the-bottom appear quite easily, the rest comes by letting the whole 
thing run in the background for a few minutes, maybe a few hours at most (never 
took up to an hour on my side).

  # WHY I THINK IT'S AN IMPORTANT ISSUE (but not a "security issue" per se) :
  The entire operating system was rendered unresponsive. While some visual cues 
appearing on the video may hint that everything isn't right, if you were 
working on something while the video plays and it crashes, you lost everything 
you were doing back to the last point you saved. I also believe - though I 
could be wrong - the the Kernel was dead as well, since REISUB doesn't work, as 
mentioned earlier.

  The bug shows it is possible for an ordinary application to cause a
  complete denial of service on the whole system. No matter what the
  problem inside Totem is, it has the power to creep down to the whole
  system (and maybe the kernel); an application running in user space
  shouldn't have such power.

  However, I struggled to find clear, easily reproducible steps that
  could result in a crash 100% of the time on any Ubuntu 18.04.4 LTS
  system. I therefore cannot guarantee it isn't just a problem on my end
  (but even if it is, I would like to understand how a bug in Totem can
  have such a huge impact).

  If my reasoning is wrong, then I apologize for the inconvenience. Feel
  free to classify my report the way it should be if need there is.

  # FILES ATTACHED :
  - 2020-04-09 01-04-59.mp4 : (I apologize for the unprofessional video... I 
tested it on a meme) A recording made with OBS of the apparition of the visual 
artifacts. The artifacts are essentially concentrated in the first minute of 
the video - the rest shows only the red/green gradients, which are still 
disturbing but much less aggressive on the eye. At the very end of the video, I 
close the target file and re-open it; fatal decision, it was at that moment 
that my computer roze and stopped responding. A few moments later, the screen 
went black, and showed the strange noise vision I described earlier and that I 
show in the other video file.
  - video20200409_011018907.mp4 : The other video file, recorded this time with 
my phone, showing the screen once it has switched to it's "noisy" state 
described earlier. There was no sound playing from the speakers, so I recorded 
with no sound (thanks Ubuntu Touch for that option). Also, my mouse froze after 
some time.
  - Screenshot from 2020-04-09 01-15-02.png : When re-playing the video file 
(which was obviously partly corrupt, since OBS crashed with the system before 
it could stop recording), I noticed the artifacts came back rather quickly. 
This time I only took a screenshot and immediately closed Totem, fearing 
another full system crash.
  - Folder "corrupt-but-playable" : It contains two video files which have 
triggered the visual artifacts.
      - y8gVYBC4jPyr8QB1.mp4 : The video you can see in the test crash.
      - 2020-04-09 01-04-59.flv : The original, "corrupt" version of the MP4 
with the same name (first in this list). Apart from having triggered the visual 
artifacts, I also notice Totem couldn't compute its duration properly (I blame 
the file for that, not Totem). Thankfully, OBS offers an option to remux 
recordings, which converted the file to an MP4 that seems clean.

  # SYSTEM INFO
  - lsb_release -rd
  Description:  Ubuntu 18.04.4
  LTS Release:  18.04

  - uname -a
  Linux desktop-u 5.3.0-46-generic #38~18.04.1-Ubuntu SMP Tue Mar 31 04:17:56 
UTC 2020 x86_64 x86_64 x86_64 GNU/Linux

  - apt-cache policy totem
  totem:
    Installed: 3.26.0-0ubuntu6.2
    Candidate: 3.26.0-0ubuntu6.2
    Version table:
   *** 3.26.0-0ubuntu6.2 500
          500 http://ca.archive.ubuntu.com/ubuntu bionic-updates/main amd64 
Packages
          100 /var/lib/dpkg/status
       3.26.0-0ubuntu6 500
          500 http://ca.archive.ubuntu.com/ubuntu bionic/main amd64 Packages

  I'm open at reproducing the bug on my computer, on request, if more
  information is needed.

  ProblemType: Bug
  DistroRelease: Ubuntu 18.04
  Package: libgstreamer1.0-0 1.14.5-0ubuntu1~18.04.1
  ProcVersionSignature: Ubuntu 5.3.0-46.38~18.04.1-generic 5.3.18
  Uname: Linux 5.3.0-46-generic x86_64
  ApportVersion: 2.20.9-0ubuntu7.14
  Architecture: amd64
  CurrentDesktop: ubuntu:GNOME
  Date: Thu Apr  9 01:23:53 2020
  InstallationDate: Installed on 2019-08-28 (224 days ago)
  InstallationMedia: Ubuntu 18.04.3 LTS "Bionic Beaver" - Release amd64 
(20190805)
  ProcEnviron:
   LANGUAGE=en_CA:en
   PATH=(custom, no user)
   XDG_RUNTIME_DIR=<set>
   LANG=en_CA.UTF-8
   SHELL=/bin/bash
  SourcePackage: gstreamer1.0
  UpgradeStatus: No upgrade log present (probably fresh install)
  XorgLog: Error: [Errno 2] No such file or directory: '/var/log/Xorg.0.log'
  --- 
  ProblemType: Bug
  ApportVersion: 2.20.9-0ubuntu7.14
  Architecture: amd64
  AudioDevicesInUse:
   USER        PID ACCESS COMMAND
   /dev/snd/controlC0:  semphris   2027 F.... pulseaudio
   /dev/snd/controlC1:  semphris   2027 F.... pulseaudio
  CurrentDesktop: ubuntu:GNOME
  DistroRelease: Ubuntu 18.04
  InstallationDate: Installed on 2019-08-28 (224 days ago)
  InstallationMedia: Ubuntu 18.04.3 LTS "Bionic Beaver" - Release amd64 
(20190805)
  MachineType: HP 260-p009
  Package: linux (not installed)
  ProcFB: 0 radeondrmfb
  ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.3.0-46-generic 
root=/dev/mapper/ubuntu--vg-root ro quiet splash vt.handoff=1
  ProcVersionSignature: Ubuntu 5.3.0-46.38~18.04.1-generic 5.3.18
  RelatedPackageVersions:
   linux-restricted-modules-5.3.0-46-generic N/A
   linux-backports-modules-5.3.0-46-generic  N/A
   linux-firmware                            1.173.17
  Tags:  bionic
  Uname: Linux 5.3.0-46-generic x86_64
  UpgradeStatus: No upgrade log present (probably fresh install)
  UserGroups: sudo vboxusers
  WifiSyslog:
   
  _MarkForUpload: True
  dmi.bios.date: 03/10/2016
  dmi.bios.vendor: AMI
  dmi.bios.version: F.03
  dmi.board.asset.tag: CNV6190N2W
  dmi.board.name: 81B4
  dmi.board.vendor: HP
  dmi.board.version: 01
  dmi.chassis.asset.tag: CNV6190N2W
  dmi.chassis.type: 3
  dmi.chassis.vendor: HP
  dmi.modalias: 
dmi:bvnAMI:bvrF.03:bd03/10/2016:svnHP:pn260-p009:pvr:rvnHP:rn81B4:rvr01:cvnHP:ct3:cvr:
  dmi.product.family: 103C_53316J G=D
  dmi.product.name: 260-p009
  dmi.product.sku: V8P62AA#ABL
  dmi.sys.vendor: HP

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1871766/+subscriptions

-- 
Mailing list: https://launchpad.net/~kernel-packages
Post to     : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help   : https://help.launchpad.net/ListHelp

Reply via email to