Did you get the hung renderer dialog? Did chrome upload a crash report?

Could you please provide the crash id from chrome://crashes for further debugging.

> Did you get the hung renderer dialog?

Sometimes.

> Did chrome upload a crash report?

Nope.

> Could you please provide the crash id from chrome://crashes for further debugging.

I will try to attain a crash id. Not sure why none has been logged yet.

Going to assign to myself with needs-feedback until I provide this info. Meanwhile, are there anything else I can do to help debug this?

If you could attach a debugger if that happens and post the stack traces from all non-trivial threads, that'd be awesome :)

Something tells me I should compile Chrome from ToT and keep using that until I reproduce the bug. Debugging Chrome Stable is useless ;-)

$ gdb -p 139453
GNU gdb (GDB) 7.9-gg19

<---snip---->

Attaching to process 139453
Reading symbols from /opt/google/chrome/chrome...done.

WARNING: no debugging symbols found in /opt/google/chrome/chrome.
Either the binary was compiled without debugging information
or the debugging information was removed (e.g., with strip or strip -g).
Debugger capabilities will be very limited.

<---snip---->

warning: Target and debugger are in different PID namespaces; thread lists and other data are likely unreliable
pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
185     ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S: No such file or directory.
(gdb) bt full
#0  pthread_cond_wait@@GLIBC_2.3.2 () at ../nptl/sysdeps/unix/sysv/linux/x86_64/pthread_cond_wait.S:185
No locals.
#1  0x00007f581105cd9a in ?? ()
No symbol table info available.
#2  0x0000328673978bb0 in ?? ()
No symbol table info available.
#3  0x0008480173d49b60 in ?? ()
No symbol table info available.
#4  0x0000328673d49ba0 in ?? ()
No symbol table info available.
#5  0x00007f58158c2a10 in ?? ()
No symbol table info available.
#6  0x0000000000000000 in ?? ()
No symbol table info available.

Stack traces attached. Anything useful in there?

Deleted: hung-tab-bt-full.txt 68.6 KB

hung-tab-bt-full.txt 68.6 KB View Download

Adding components based on stack trace.

thanks for the stacktrace!

Assigning to alokp@ who last touched the Pipeline::Stop call

Likely this means the audio stack is hung in some way. Generally when we see this on Linux it means there's a misconfiguration of pulse. If you check your syslog do you see any audio errors?

Sometimes executing sudo pulseaudio -k from the command line will spit these out to the console.

Yep, pulseaudio prints to syslog:

Feb  6 09:49:33 rouslan pulseaudio[8610]: [pulseaudio] module-alsa-card.c: Failed to find a working profile.
Feb  6 09:49:33 rouslan pulseaudio[8610]: [pulseaudio] module.c: Failed to load module "module-alsa-card" (argument: "device_id="2" name="usb-046d_08ca_0110DFA6-02-U0x46d0x8ca" card_name="alsa_card.usb-046d_08ca_0110DFA6-02-U0x46d0x8ca" namereg_fail=false tsched=yes fixed_latency_range=no ignore_dB=no deferred_volume=yes use_ucm=yes card_properties="module-udev-detect.discovered=1""): initialization failed.

Let me attempt to fix pulseaudio config, see how many hangs I get the following week, and then get back to you.

Fiing pusleaudio config did the trick.