[snap] chromium crashes regularly

Bug #1960313 reported by Chris Guiver
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
chromium-browser (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

This was initially mentioned here (https://discourse.ubuntu.com/t/call-for-testing-chromium-browser-deb-to-snap-transition/11179/317)

** Description

For awhile now I just find my chromium just closes and is gone. It’s been at least the last two updates (I get the nag screen on my screen telling me an update is available; I tend to ignore it but reboot my box every ~fortnight so it’ll update at least once per fortnight as I ensure it’s upgraded before I load it after reboot).

It’s ~10:40 local time, and I’ve lost the chromium snap at least twice today… no clues as to why, just I get to see my wallpaper on that part of my monitor as the window silently disappears.

On re-opening I get a ‘closed ; do you want to restore you tabs’ dialog (something like that anyway) which works, but my box has been up ~13 days, and it was giving me trouble before this reboot as I hoped that reboot would help; alas no; neither last update(s).

If I was to guess how long, I'd guess 20-30 days if not more; ie. I'm basing that on two ~14 day reboot cycles... I can't recall if started during the last cycle or just prior to my second-last reboot ~28 days ago

As I file this bug; it's now ~23:15 and whilst I mentioned it'd closed twice prior to ~10:40 (I turn box on ~07:30), I'm pretty sure the window has not closed since I posted on the community hub. I'm not usually using the window when it closes; though its closed ~twice whilst I'm using it.

** Limited details sorry

I'm not very specific here sorry, I'd file it as 'incomplete', but I don't know how/where to get details. As a snap there is nothing in /var/crash; so as an aside; is there somewhere I can look for clues as to why etc. (like the .crash file in /var/crash, even if I can only understand bits of it)

My box is slow (2009 dell desktop)... and I've got in the habit of keeping a terminal open with `htop` running so I can detect if `firefox` (deb package still) or `chromium` are losing RAM.. I've concluded (rightly/wrongly) it's extension related; but if I recognize extensive swapping I look at the `htop` window and if either browser is using a load more RAM than usual & swap size is growing, I tend to close browser, wait for OS to recover the RAM, then re-open.. I've NOT been doing this with chromium of late; so I wondered if it's started crashing where I'd normally notice a slowing box & switch to my htop window, then close the window down... HOWEVER it could also be the regular crashing means the 'ram bleed' issue isn't getting bad enough for the thrashing to become a problem; it's usually only a problem after 2-4 days.. I'm convinced the issue (what I call 'ram bleed') is related to extensions, as the issue didn't occur with no extensions being used; but I prefer having privacy badger/ublock running
- google docs offline
- privacy badger
- ublock origin
- startpage (this was only ~recently added; not this one)
FYI: I use same extensions with firefox too (except no startpage), and issue is the same; ~same frequency too (2-4 days)

Anyway; if you can tell me if it'd be helpful to get/keep certain details; I'll do it, as I realize I'm not giving much.

ProblemType: Bug
DistroRelease: Ubuntu 22.04
Package: chromium-browser 1:85.0.4183.83-0ubuntu2
ProcVersionSignature: Ubuntu 5.15.0-17.17-generic 5.15.12
Uname: Linux 5.15.0-17-generic x86_64
ApportVersion: 2.20.11-0ubuntu76
Architecture: amd64
CasperMD5CheckResult: unknown
CurrentDesktop: LXQt
DRM.card0-DVI-I-1:
 enabled: enabled
 dpms: On
 status: connected
 edid-base64: AP///////wAs7hAkAQEBAQAaAQOANB14KrclpFdRoCYQUFS/74CzAKlAlQCBwIGAqcBxT4EAAjqAGHE4LUBYLEUACSUhAAAeZiFQsFEAGzBAcDYACSUhAAAeAAAA/QA3TB5SEQAKICAgICAgAAAA/ABLQUxFRDI0TU9OQ0EKAJo=
 modes: 1920x1080 1600x1200 1680x1050 1600x900 1280x1024 1280x1024 1440x900 1360x768 1280x800 1152x864 1280x720 1024x768 1024x768 1024x768 832x624 800x600 800x600 800x600 800x600 640x480 640x480 640x480 640x480 720x400
DRM.card0-DVI-I-2:
 enabled: enabled
 dpms: On
 status: connected
 edid-base64: AP///////wAEchMBz9ogAhYUAQOAMx14KsKFpFZNnCUSUFSzDABxT4EAgYCVALMA0cABAQEBAjqAGHE4LUBYLEUA/h8RAAAeAAAA/QA4Sx5TEQAKICAgICAgAAAA/ABBY2VyIEcyMzVICiAgAAAA/wBMSkswVzAxNjQzMjAKAIg=
 modes: 1920x1080 1680x1050 1280x1024 1440x900 1280x800 1152x864 1024x768 1024x768 800x600 800x600 640x480 640x480 720x400
Date: Tue Feb 8 22:48:11 2022
DiskUsage:
 Filesystem Type Size Used Avail Use% Mounted on
 /dev/sda4 xfs 32G 32G 63M 100% /home
 tmpfs tmpfs 3.9G 174M 3.7G 5% /dev/shm
 /home/guiverc/.Private ecryptfs 32G 32G 63M 100% /home/guiverc
EcryptfsInUse: Yes
InstallationDate: Installed on 2017-11-16 (1545 days ago)
InstallationMedia: Ubuntu 17.10 "Artful Aardvark" - Release amd64 (20171018)
MachineType: Dell Inc. OptiPlex 960
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-5.15.0-17-generic root=UUID=afa7971f-3dd5-4b30-9c98-0af3e56a6f2b ro quiet splash vt.handoff=7
RebootRequiredPkgs: Error: path contained symlinks.
Snap.Changes:
 ID Status Spawn Ready Summary
 760 Done 2022-02-07T23:47:32+11:00 2022-02-07T23:50:10+11:00 Refresh snap "chromium"
Snap.ChromeDriverVersion: Error: command ['snap', 'run', 'chromium.chromedriver', '--version'] failed with exit code 1: cannot open path of the current working directory: Permission denied
Snap.ChromiumVersion: Error: command ['snap', 'run', 'chromium', '--version'] failed with exit code 1: cannot open path of the current working directory: Permission denied
SourcePackage: chromium-browser
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 12/05/2011
dmi.bios.release: 10.0
dmi.bios.vendor: Dell Inc.
dmi.bios.version: A14
dmi.board.name: 0F428D
dmi.board.vendor: Dell Inc.
dmi.board.version: A00
dmi.chassis.asset.tag: 1701283234
dmi.chassis.type: 3
dmi.chassis.vendor: Dell Inc.
dmi.modalias: dmi:bvnDellInc.:bvrA14:bd12/05/2011:br10.0:svnDellInc.:pnOptiPlex960:pvr:rvnDellInc.:rn0F428D:rvrA00:cvnDellInc.:ct3:cvr:sku:
dmi.product.name: OptiPlex 960
dmi.sys.vendor: Dell Inc.

Revision history for this message
Chris Guiver (guiverc) wrote :
Revision history for this message
Olivier Tilloy (osomon) wrote :

Could you please share the output of the following commands?

    snap info --abs-time gnome-3-38-2004

    snap info --abs-time core20

    snap get core experimental.refresh-app-awareness

summary: - chromium-browser has started closing regularly
+ [snap] chromium crashes regularly
Changed in chromium-browser (Ubuntu):
status: New → Incomplete
Revision history for this message
Chris Guiver (guiverc) wrote :

guiverc@d960-ubu2:/de2900/others/railway$ snap info --abs-time gnome-3-38-2004
name: gnome-3-38-2004
summary: Shared GNOME 3.38 Ubuntu stack
publisher: Canonical✓
store-url: https://snapcraft.io/gnome-3-38-2004
license: unset
description: |
  This snap includes a GNOME 3.38 stack (the base libraries and desktop
  integration components) and shares it through the content interface.
snap-id: rw36mkAjdIKl13dzfwyxP87cejpyIcct
tracking: latest/stable
refresh-date: 2022-01-11T15:21:38+11:00
channels:
  latest/stable: 0+git.cd626d1 2021-11-22T21:01:02Z (87) 259MB -
  latest/candidate: 0+git.1f9014a 2022-01-08T01:54:11Z (99) 260MB -
  latest/beta: ↑
  latest/edge: ↑
installed: 0+git.cd626d1 (87) 259MB -

---

guiverc@d960-ubu2:/de2900/others/railway$ snap info --abs-time core20
name: core20
summary: Runtime environment based on Ubuntu 20.04
publisher: Canonical✓
store-url: https://snapcraft.io/core20
contact: https://github.com/snapcore/core20/issues
license: unset
description: |
  The base snap based on the Ubuntu 20.04 release.
type: base
snap-id: DLqre5XGLbDqg9jPtiAhRRjDuPVa5X1q
tracking: latest/stable
refresh-date: 2022-01-27T12:47:40+11:00
channels:
  latest/stable: 20220114 2022-01-28T16:10:53Z (1328) 64MB -
  latest/candidate: 20220114 2022-01-25T21:05:26Z (1328) 64MB -
  latest/beta: 20220207 2022-02-07T17:05:29Z (1340) 64MB -
  latest/edge: 20220207 2022-02-07T12:16:10Z (1340) 64MB -
installed: 20220114 (1328) 64MB base

---

guiverc@d960-ubu2:/de2900/others/railway$ snap get core experimental.refresh-app-awareness
error: access denied (try with sudo)
guiverc@d960-ubu2:/de2900/others/railway$ sudo snap get core experimental.refresh-app-awareness
[sudo] password for guiverc:
true

Revision history for this message
Olivier Tilloy (osomon) wrote :

Thanks for the additional info.
We can rule out a background refresh of chromium or any of its dependencies (core20, gnome-3-38-2004, gtk-common-themes).

Since the crash reporter isn't enabled in chromium (it's a chrome-specific thing), may I suggest that you run chromium from a terminal with verbose logging enabled, and next time a crash happens share the relevant output here?

    snap run chromium --enable-logging=stderr --v=1

Changed in chromium-browser (Ubuntu):
status: Incomplete → New
Revision history for this message
Chris Guiver (guiverc) wrote (last edit ):
Download full text (7.4 KiB)

Sorry, I missed the prior crash, but hopefully this one contains something that is useful

https://pastebin.ubuntu.com/p/c5TTS6bXdP/

Not sure if helpful, but I use `chromium` for sites related to google (or where I use google SSO to login).

- gmail (regularly refresh this; it's set for manual refresh though so I'm hitting 'inbox' to refresh)
- google docs (Ubuntu News etc; not today)
- mewe (not often, ; I posted UWN this morning; ~7 hours ago)
- youtube (my primary use of chromium; about all today really)

chromium is left open & idle; most crashes are whilst idle.

this time it wasn't idle; I was refreshing gmail page on one window, and in process of switching to a different youtube vid on other window..

(most sites (non-google); including launchpad etc. I use firefox)

last few lines

[260919:260923:0215/170900.786576:VERBOSE1:tls_handshaker.cc(104)] TlsClient: Continuing handshake
[260919:260923:0215/170900.787131:VERBOSE1:tls_handshaker.cc(115)] TlsClient: SSL_do_handshake returned when entering early data. After retry, rv=-1, SSL_in_early_data=1
[260919:260923:0215/170900.886186:VERBOSE1:tls_handshaker.cc(104)] TlsClient: Continuing handshake
[260919:260923:0215/170900.886515:VERBOSE1:tls_handshaker.cc(104)] TlsClient: Continuing handshake
[260919:260923:0215/170900.886602:VERBOSE1:tls_client_handshaker.cc(513)] Client: handshake finished
[260919:260923:0215/170901.181857:VERBOSE1:network_delegate.cc(34)] NetworkDelegate::NotifyBeforeURLRequest: https://signaler-pa.clients6.google.com/punctual/multi-watch/channel?VER=8&gsessionid=gWmA3cjvRDO-Io5thTAHCcfnERDeMKLhwP9a777x4MI&key=AIzaSyAWGrfCCr7albM3lmCc937gx4uIphbpeKQ&RID=rpc&SID=yEQaMB5z0Mt6Fvhpwnd8gQ&CI=0&AID=44&TYPE=xmlhttp&zx=258pfetv9dc5&t=1
[261888:1:0215/171008.533492:VERBOSE1:navigator_beacon.cc(88)] Send a beacon to "https://www.youtube.com/api/stats/qoe?fmt=134&afmt=251&cpn=WLB0NxJoDpXAjfpi&el=detailpage&ns=yt&fexp=23858057%2C23983296%2C23986027%2C24001373%2C24002022%2C24002025%2C24002923%2C24004644%2C24007246%2C24044261%2C24080738%2C24082662%2C24135310%2C24139174%2C24148482%2C24153108%2C24154238%2C24166123%2C24170254&cl=427307184&seq=2988&docid=IOEtbGcWoos&ei=gWEHYvyMCcnT3LUP6fi20AI&event=streamingstats&plid=AAXXzR-NuRq4yiwf&referrer=https%3A%2F%2Fwww.youtube.com%2F&sdetail=p%3A%2F&sourceid=y&cbr=Chrome&cbrver=98.0.4758.80&c=WEB&cver=2.20220211.01.00&cplayer=UNIPLAYER&cos=X11&cplatform=DESKTOP&vps=178983.821:B&bwe=178983.821:544511&bat=178983.821:1:1&cmt=178983.821:0.000&bh=178983.821:0.000", initiator = "https://www.youtube.com/watch?v=IOEtbGcWoos"
[261673:1:0215/171008.533581:VERBOSE1:navigator_beacon.cc(88)] Send a beacon to "https://www.youtube.com/api/stats/qoe?fmt=396&afmt=251&cpn=9cr_pXd_iSq_jENs&el=detailpage&ns=yt&fexp=23858057%2C23983296%2C23986027%2C24001373%2C24002022%2C24002025%2C24002923%2C24004644%2C24007246%2C24044261%2C24080738%2C24082662%2C24135310%2C24139174%2C24148482%2C24153108%2C24154238%2C24166123%2C24170254&cl=427307184&seq=2988&docid=WyvnnCTaYX8&ei=RmEHYsDiG5j_4-EPzfq_sAo&event=streamingstats&plid=AAXXzRwOLTlBdJbO&cbr=Chrome&cbrver=98.0.4758.80&c=WEB&cver=2.20220211.01.00&cplayer=UNIPLAYER&cos=X11&cplatform=DESKTOP&vps=179036.589:S&bwe=179036.589...

Read more...

Revision history for this message
Olivier Tilloy (osomon) wrote :

Thanks for the additional info and for the logs Chris. So Chromium was abruptly killed, and the logs don't contain any hint as to why this might have happened.

I wonder whether this could be an out of memory situation, where the kernel decided to kill Chromium to free up memory. Could you inspect the output of `journalctl` around the timestamp when the crash occurred?

In the DiskUsage section of the data attached by apport, I'm seeing that your /home partition is almost full (63M free out of 32GB), maybe that could explain this sort of abrupt termination if Chromium is trying to write profile or cache data to disk?

Revision history for this message
Chris Guiver (guiverc) wrote :

I'm embarrassed now as I'd not detected that (and further telegram was also shutdown & I'd not even noticed..)

Alas I'm somewhat aware of disk space issues.. (but deal with that only if the box won't let me login) but as I'd not been denied login for awhile hadn't noticed how serious it'd become..

You were spot on - OOM errors (chromium, then telegram)

THANK YOU FOR THE SUPPORT :)

& sorry I'd not picked the issue & my issues.

Feb 15 17:10:20 d960-ubu2 systemd-oomd[949]: Killed /user.slice/user-1000.slice/user@1000.service/app.slice/snap.chromium.chromium.82bbbfd7-c79b-4b57-b425-e543631ca054.scope due to memory p>
Feb 15 17:10:22 d960-ubu2 mtp-probe[446025]: checking bus 5, device 89: "/sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2"
Feb 15 17:10:23 d960-ubu2 systemd[5610]: snap.chromium.chromium.82bbbfd7-c79b-4b57-b425-e543631ca054.scope: Consumed 5h 29min 24.000s CPU time.
Feb 15 17:10:22 d960-ubu2 mtp-probe[446025]: bus: 5, device: 89 was not an MTP device
Feb 15 17:10:23 d960-ubu2 upowerd[151570]: treating change event as add on /sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2
Feb 15 17:10:23 d960-ubu2 mtp-probe[446048]: checking bus 5, device 89: "/sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2"
Feb 15 17:10:23 d960-ubu2 mtp-probe[446048]: bus: 5, device: 89 was not an MTP device
Feb 15 17:10:34 d960-ubu2 kernel: usb 5-2: input irq status -75 received
Feb 15 17:10:34 d960-ubu2 kernel: usb 5-2: USB disconnect, device number 89
Feb 15 17:10:35 d960-ubu2 kernel: usb 5-2: new low-speed USB device number 90 using uhci_hcd
Feb 15 17:10:35 d960-ubu2 kernel: usb 5-2: New USB device found, idVendor=0461, idProduct=4d15, bcdDevice= 2.00
Feb 15 17:10:35 d960-ubu2 kernel: usb 5-2: New USB device strings: Mfr=0, Product=2, SerialNumber=0
Feb 15 17:10:35 d960-ubu2 kernel: usb 5-2: Product: USB Optical Mouse
Feb 15 17:10:35 d960-ubu2 kernel: input: USB Optical Mouse as /devices/pci0000:00/0000:00:1a.2/usb5/5-2/5-2:1.0/0003:0461:4D15.01CF/input/input499
Feb 15 17:10:35 d960-ubu2 kernel: hid-generic 0003:0461:4D15.01CF: input,hidraw0: USB HID v1.11 Mouse [USB Optical Mouse] on usb-0000:00:1a.2-2/input0
Feb 15 17:10:35 d960-ubu2 upowerd[151570]: treating change event as add on /sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2
Feb 15 17:10:35 d960-ubu2 mtp-probe[446683]: checking bus 5, device 90: "/sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2"
Feb 15 17:10:35 d960-ubu2 mtp-probe[446683]: bus: 5, device: 90 was not an MTP device
Feb 15 17:10:35 d960-ubu2 mtp-probe[446698]: checking bus 5, device 90: "/sys/devices/pci0000:00/0000:00:1a.2/usb5/5-2"
Feb 15 17:10:35 d960-ubu2 mtp-probe[446698]: bus: 5, device: 90 was not an MTP device
Feb 15 17:10:36 d960-ubu2 systemd-oomd[949]: Killed /user.slice/user-1000.slice/user@1000.service/app.slice/snap.telegram-desktop.telegram-desktop.c0a93946-0960-44ed-aeb3-f9903b01056b.scope>
Feb 15 17:10:36 d960-ubu2 systemd[5610]: snap.telegram-desktop.telegram-desktop.c0a93946-0960-44ed-aeb3-f9903b01056b.scope: Consumed 49min 26.108s CPU time.
Feb 15 17:10:57 d960-ubu2 kernel: usb 5-2: reset low-speed USB device number 90 using uhci_hcd

Changed in chromium-browser (Ubuntu):
status: New → Invalid
Revision history for this message
Olivier Tilloy (osomon) wrote :

You're welcome, I'm glad we eventually figured it out.

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.