You are not logged in.
Pages: 1
Hi all,
After a Discover update notification and implementation 30 minutes ago, Chromium is not opening and neither is kWalletManager automatically. I can open the latter and sign in but this does not help Chromium.
Each time I try to open Chromium, it registers in Crashed Processes Viewer, but I don't seem to be able to make any improvements via the Interactive Debugger.
I'm running Q4OS Plasma on a Macbook Pro 2013, installed in December 2025 and updated at each suggestion. Running has been flawless up to now. I'm very anxious to fix this as I am tackling Harvard's CS50 course and am in the middle of the week 5 problem set! And I really enjoy all of Q4OS's many aspects.
I'm logging on here in Chrome on a Windows 11 machine.
Your suggestions are welcome.
Many thanks,
philc
Last edited by q4osteam (2026-07-09 10:23)
Offline
Crashed process trace:
PID: 2897 (chromium)
UID: 1000 (philc)
GID: 1000 (philc)
Signal: 5 (TRAP)
Timestamp: Sun 2026-07-05 15:02:31 BST (31min ago)
Command Line: $'/usr/lib/chromium/chromium --wm-window-animations-disabled --show-component-extension-options --enable-gpu-rasterization --no-default-browser-check --disable-pings --media-router=0 --enable-remote-extensions --load-extension'
Executable: /usr/lib/chromium/chromium
Control Group: /user.slice/user-1000.slice/user@1000.service/app.slice/app-chromium@1d111f0c83ab48dc929311540a33dfb0.service
Unit: user@1000.service
User Unit: app-chromium@1d111f0c83ab48dc929311540a33dfb0.service
Slice: user-1000.slice
Owner UID: 1000 (philc)
Boot ID: 9b1bbb8d127c40cbb19da983eb141d6b
Machine ID: e792170635c64b128013bceff3642263
Hostname: philip-macbookpro112
Storage: /var/lib/systemd/coredump/core.chromium.1000.9b1bbb8d127c40cbb19da983eb141d6b.2897.1783260151000000.zst (present)
Size on Disk: 5.1M
Message: Process 2897 (chromium) of user 1000 dumped core.
Module libblkid.so.1 from deb util-linux-2.41-5.amd64
Module libsystemd.so.0 from deb systemd-257.13-1~deb13u1.amd64
Module libmount.so.1 from deb util-linux-2.41-5.amd64
Module libstdc++.so.6 from deb gcc-14-14.2.0-19.amd64
Module libatomic.so.1 from deb gcc-14-14.2.0-19.amd64
Module libgcc_s.so.1 from deb gcc-14-14.2.0-19.amd64
Module libudev.so.1 from deb systemd-257.13-1~deb13u1.amd64
Module libzstd.so.1 from deb libzstd-1.5.7+dfsg-1.amd64
/* Stack traces removed */
Last edited by philc (2026-07-08 14:29)
Offline
This is the output on running Chromium in Konsole:
MESA-INTEL: warning: Haswell Vulkan support is incomplete
libva error: /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so init failed
[0705/163537.567476:ERROR:third_party/crashpad/crashpad/snapshot/elf/elf_dynamic_a
rray_reader.h:64] tag not found
Trace/breakpoint trap (core dumped)
Offline
We are not able to reproduce the issue, Chromium starts fine here. So it seems to be hardware dependent issue. We attach a few AI tips, some of them may help. Please try and post back results.
You could also try to install Firefox or Google Chrome from the Q4OS software center.
----
AI generated tips:
From the crash report, a few things stand out:
* Chromium **starts but immediately aborts with `SIGTRAP` (Trace/breakpoint trap)**.
* The console shows:
```
MESA-INTEL: warning: Haswell Vulkan support is incomplete
libva error: ... iHD_drv_video.so init failed
...
Trace/breakpoint trap (core dumped)
```
* This happened **immediately after a system update**, and **kWalletManager also stopped auto-starting**, suggesting the update changed a shared library or graphics component rather than Chromium itself.
The Vulkan and VA-API warnings are usually **warnings**, not fatal errors. The `SIGTRAP` strongly suggests Chromium is deliberately aborting because of an internal check failing—often due to a bad graphics stack, incompatible library, or packaging bug.
## First thing to try: disable GPU
From a terminal:
```bash
chromium --disable-gpu
```
If that works, the update likely introduced a Mesa/Vulkan regression.
If it starts, you can then try:
```bash
chromium --disable-vulkan
```
or permanently disable Vulkan in Chromium's flags.
---
## Try a clean profile
Sometimes an update changes the profile format.
```bash
mv ~/.config/chromium ~/.config/chromium.old
chromium
```
If Chromium starts, the browser itself is fine and the profile is the culprit.
---
## Check whether today's update upgraded Mesa
Run
```bash
apt history
```
or
```bash
grep " upgrade " /var/log/apt/history.log
```
and look for packages such as
* mesa
* libdrm
* chromium
* libva
* intel-media-driver
If Chromium and Mesa were upgraded together, that is a strong clue.
---
## Force Chromium to ignore VA-API
The
```
libva error: iHD_drv_video.so init failed
```
may indicate the Intel media driver is now incompatible.
Try
```bash
LIBVA_DRIVER_NAME=i965 chromium
```
or
```bash
chromium --disable-accelerated-video-decode
```
Haswell machines sometimes behave better with the older `i965` VA driver than the newer `iHD` driver.
---
## Get a symbolic stack trace
The current backtrace contains only raw addresses, which isn't enough to identify the failure.
Please run:
```bash
coredumpctl info chromium
```
and especially
```bash
coredumpctl gdb chromium
```
Inside GDB, enter:
```
bt
```
That should reveal the function where Chromium is trapping.
---
## Check whether this is a packaging bug
Please also post the output of:
```bash
chromium --version
```
and
```bash
dpkg -l chromium
```
along with
```bash
cat /etc/os-release
```
Since this started immediately after a Discover update on a Haswell-based 2013 MacBook Pro, my suspicion is:
1. a Chromium package regression,
2. a Mesa/Vulkan regression,
3. or an Intel media-driver (`libva`) incompatibility introduced by the update.
Those are considerably more likely than filesystem corruption or a damaged user profile.
If you can provide the outputs of:
```bash
chromium --version
apt history
coredumpctl info chromium
```
I can usually narrow it down to the specific package responsible.
Offline
Hi, thank you,
This is some of the core dump information you requested:
--show-component-extension-options\ --enable-gpu-rasterization\ --no-default-brow
ser-check\ --disable-pings\ --media-router=0\ --enable-remote-extensions\ --load-e
xtension --show-component-extension-options\ --enable-gpu-rasterization\ --no-defa
ult-browser-check\ --disable-pings\ --media-router=0\ --enable-remote-extensions\
--load-extension --enable-gpu-rasterization\ --no-default-browser-check\ --disable
-pings\ --media-router=0\ --enable-remote-extensions\ --load-extension --no-defaul
t-browser-check\ --disable-pings\ --media-router=0\ --enable-remote-extensions\ --
load-extension --disable-pings\ --media-router=0\ --enable-remote-extensions\ --lo
ad-extension --media-router=0\ --enable-remote-extensions\ --load-extension --enab
le-remote-extensions\ --load-extension --load-extension'.
Program terminated with signal SIGTRAP, Trace/breakpoint trap.
#0 0x0000559ee034edb7 in ?? ()
[Current thread is 1 (Thread 0x7f3b55f63300 (LWP 8935))]
(gdb)
None of the graphics bypasses worked and I got the same messages in Konsole.
Resetting the profile folder did not work.
Chromium version: Chromium 150.0.7871.46 built on GNU/Linux 13 (trixie)
Information from /var/log/apt/history.log:
Start-Date: 2026-07-05 14:35:04
Commandline: packagekit role='update-packages'
Requested-By: philc (1000)
Upgrade: chromium-sandbox:amd64 (149.0.7827.196-1~deb13u1, 150.0.7871.46-1~deb13u1
), chromium:amd64 (149.0.7827.196-1~deb13u1, 150.0.7871.46-1~deb13u1), chromium-l1
0n:amd64 (149.0.7827.196-1~deb13u1, 150.0.7871.46-1~deb13u1), chromium-common:amd6
4 (149.0.7827.196-1~deb13u1, 150.0.7871.46-1~deb13u1)
End-Date: 2026-07-05 14:35:12
Google AI opinion: Only the Chromium packages updated to version 150.0.7871.46 at 14:35. No Mesa or graphics drivers changed, confirming a direct package regression with Chromium 150 on Haswell hardware.
I hope this is of assistance.
philc
Offline
Please try the tips by AI below. The most related by our opinion would be:
$ chromium --disable-extensions
and
"Test with a completely new user"
see below.
----
AI generated:
The important points are:
* **Only Chromium packages were upgraded** (149 → 150).
* No Mesa, libdrm, libva, Intel driver, or kernel packages were updated.
* A **new profile** behaves identically.
* `--disable-gpu`, `--disable-vulkan`, etc. make no difference.
* The crash occurs immediately with **SIGTRAP**, before any useful backtrace is produced.
That combination makes me think this is **very likely a Chromium 150 regression** rather than anything wrong with your Q4OS installation.
### One thing I'd still like to check
The GDB session stopped here:
```
#0 0x0000559ee034edb7 in ?? ()
(gdb)
```
Could you type:
```
bt
```
or preferably
```
thread apply all bt
```
Sometimes GDB can resolve symbols that weren't shown in the initial stop message.
If it still shows only `??`, that's actually further evidence that Chromium intentionally triggered the trap before meaningful symbols could be produced.
## I'd now try two more tests
### 1. Disable extensions completely
Run
```bash
chromium --disable-extensions
```
or
```bash
chromium --disable-extensions --disable-component-extensions-with-background-pages
```
The command line in the core dump shows Chromium loading component extensions. If one of those is triggering the assertion, this may get past it.
---
### 2. Test with a completely new user
Create a temporary user:
```bash
sudo adduser testchrome
```
Log into Plasma as that user and launch Chromium.
If it crashes there too, you've essentially proved it's a system-wide packaging bug.
---
## The package history is the strongest evidence
This:
```
149.0.7827.196
→
150.0.7871.46
```
with absolutely nothing else changing is about as clean an A/B test as you can get.
If version 149 worked five minutes before and 150 crashes immediately afterwards—even with a new profile—that strongly implicates Chromium 150 itself.
## I would also try running without the sandbox
Just as a diagnostic:
```bash
chromium --no-sandbox
```
I **don't recommend using this permanently**, but if it suddenly starts, that points to a sandbox regression.
## If you need Chromium immediately
You have a few options:
* install Firefox temporarily so you can continue your CS50 coursework;
* downgrade to Chromium 149 if the package is still available in your package cache;
* wait for Debian/Q4OS to publish a fixed Chromium 150 build if this proves to be an upstream regression.
One question: **does Firefox (or another Chromium-based browser such as Brave, Edge, or Vivaldi, if installed) still work on this machine?** The answer to that will help determine whether this is specific to the Debian Chromium package or affects all Chromium-based browsers on your system.
Offline
I can confirm that after update Chromium does not start on my system running Q4OS Plasma—or rather, it only starts with an empty profile (chromium --temp-profile).
Offline
The problem has now been classified as a serious bug in Debian:
Offline
@maroc
+1 Thanks for the info.
@philc
You could also try to run Chromium within x11 session instead of wayland.
Offline
Yes, we are able to reproduce the issue now. It's not Q4OS, but Debian upstream one https://bugs.debian.org/cgi-bin/bugrepo … ug=1141488 . We are waiting for the fix by Debian developers.
Offline
I confirm that I'm up and running on Firefox ESR ![]()
Offline
Is there a way to revert to/reinstall the previous version?
Offline
According to https://bugs.debian.org/cgi-bin/bugrepo … ug=1141488 the bug is now fixed for Debian/Q4OS so you only need to update from repositories:
$ sudo apt update
$ sudo apt -V full-upgrade
optionally reboot
Buggy version: chromium/150.0.7871.46-1~deb13u1, chromium/150.0.7871.46-1
Fixed version: chromium/150.0.7871.100-1
Offline
Pages: 1