You are not logged in.

#1 2014-10-21 18:06:02

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

System not resuming from suspending to RAM correctly (pulseaudio)

Hi,

I've been having this issue where after I resume the system from a suspend to RAM some things will not work properly. (e.g., gmrun will take ages to open the text box, guake is not able to open new terminal windows, thunar will break randomly, etc.)

Some of the behaviour leads me to believe this might be related to pulseaudio not resuming properly: the server stops responding after resuming (pasuspender will timeout or get connection refused, -k won't do anything, kill -9 makes the process go zombie, alsamixer won't open most of the times, and when it does it's not associated with pulseaudio anymore, if falls back to my sound card).

This has become specially annoying because it started persisting after reboots until pulseaudio decides to start working again...

Furthermore, if I resume from RAM and try to shutdown/reboot the computer, I get an error (see image below) and the computer will just freeze.

error

I don't even know where to start troubleshooting this one... although the error does look like it might be related to trying to hot-add hardware upon waking from RAM.

Cheers!

Offline

#2 2014-10-22 02:42:16

Pse
Member
Registered: 2008-03-15
Posts: 415

Re: System not resuming from suspending to RAM correctly (pulseaudio)

What makes you think this might be related to Pulseaudio? Have you tried uninstalling it to confirm it is indeed the cause?

Offline

#3 2014-10-23 09:40:41

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

Re: System not resuming from suspending to RAM correctly (pulseaudio)

the behaviour I described made me think it was pulseaudio. following your lead, I uninstalled Pulseaudio and the problem has stopped... hmm

Offline

#4 2014-10-24 08:21:25

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

Re: System not resuming from suspending to RAM correctly (pulseaudio)

I would appreciate some help troubleshooting this, as pulseaudio is a requirement for lots of software nowadays, thanks!

Offline

#5 2014-10-24 09:29:08

V1del
Forum Moderator
Registered: 2012-10-16
Posts: 25,265

Re: System not resuming from suspending to RAM correctly (pulseaudio)

You'll have to find the real error, the screenshot you posted has nothing to do with pulseaudio (or a system freeze for that matter), this happens if you have usb 3.0 ports occupied by usb2.0 devices as the kernel has to fallback to an usb2 driver despite the port theoretically using usb3, it is normal to some extent. nothing in dmesg? is there any indication when starting pulse with

pulseaudio -vvvvv

? you have to provide more information

Last edited by V1del (2014-10-24 09:35:45)

Online

#6 2014-10-26 13:15:55

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

Re: System not resuming from suspending to RAM correctly (pulseaudio)

you're right the error is not associated with pulseaudio nor the freeze when shuting down

I have installed and started pulseaudio with:

pulseaudio -vvvvv

as you suggested, I couldn't find anything in particular, but here's the output: http://pastebin.com/ZSAS4b20

I put the computer to sleep a couple of times, but there were no symptoms so my guess is this might be related to starting pulse automatically (with 'start-pulseaudio-x11' in the .desktop file) or it will be triggered after I reboot.

Offline

#7 2014-10-26 16:17:30

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

Re: System not resuming from suspending to RAM correctly (pulseaudio)

ok, so after the last test I ^C the process, and even got a message saying the daemon had been terminated. However I started having the same symptons again.

If I try to start pulseaudio again, I get the error that the daemon is already running: http://pastebin.com/cGKLbzvX

So I'm guessing this has to do with the daemon not being terminated properly.

Offline

#8 2014-10-26 16:37:09

Stebalien
Member
Registered: 2010-04-27
Posts: 1,239
Website

Re: System not resuming from suspending to RAM correctly (pulseaudio)

saz wrote:

ok, so after the last test I ^C the process, and even got a message saying the daemon had been terminated. However I started having the same symptons again.

If I try to start pulseaudio again, I get the error that the daemon is already running: http://pastebin.com/cGKLbzvX

So I'm guessing this has to do with the daemon not being terminated properly.

Have you checked (e.g. using ps)? If so, try attaching gdb (or even just strace) to the process to see what it is doing.


Steven [ web : git ]
GPG:  327B 20CE 21EA 68CF A7748675 7C92 3221 5899 410C

Offline

#9 2014-11-09 22:56:03

saz
Member
From: Lisbon
Registered: 2008-04-19
Posts: 115

Re: System not resuming from suspending to RAM correctly (pulseaudio)

yes ps does see the process, but if I try to kill it it goes zombie.

I couldn't make much out of gdb nor strace. but I do think I might have found a clue on what's happening.

before I suspend the computer: (note the card indexes jump from 0 to 2)

%  aplay -l | grep card
card 0: PCH [HDA Intel PCH], device 0: 92HD91BXX Analog [92HD91BXX Analog]
card 2: HDMI_1 [HDA Intel HDMI], device 3: HDMI 0 [HDMI 0]
card 2: HDMI_1 [HDA Intel HDMI], device 7: HDMI 1 [HDMI 1]
card 2: HDMI_1 [HDA Intel HDMI], device 8: HDMI 2 [HDMI 2

After the computer wakes up from sleep:

%  aplay -l | grep card
card 0: PCH [HDA Intel PCH], device 0: 92HD91BXX Analog [92HD91BXX Analog]
card 1: HDMI [HDA ATI HDMI], device 3: HDMI 0 [HDMI 0]
card 1: HDMI [HDA ATI HDMI], device 7: HDMI 1 [HDMI 1]
card 2: HDMI_1 [HDA Intel HDMI], device 3: HDMI 0 [HDMI 0]
card 2: HDMI_1 [HDA Intel HDMI], device 7: HDMI 1 [HDMI 1]
card 2: HDMI_1 [HDA Intel HDMI], device 8: HDMI 2 [HDMI 2]

So two new devices for a card 1 (which didn't even show up before) appeared.

To be honest I just need the Intel PCH audio card (card 0) so I thought I'd block the other ones from being loaded.

% lsmod | grep -i snd                              [~] | 0
snd_hda_codec_idt      56952  1            
snd_hda_codec_generic    63126  3 snd_hda_codec_idt
snd_hda_intel          26387  4 
snd_hda_controller     26938  1 snd_hda_intel
snd_hda_codec         108536  4 snd_hda_codec_idt,snd_hda_codec_generic,snd_hda_intel,snd_hda_controller
snd_hwdep              17244  1 snd_hda_codec
snd_pcm                88487  3 snd_hda_codec,snd_hda_intel,snd_hda_controller
snd_timer              26614  1 snd_pcm
snd                    73436  15 snd_hwdep,snd_timer,snd_hda_codec_idt,snd_pcm,snd_hda_codec_generic,snd_hda_codec,snd_hda_intel
soundcore              13031  2 snd,snd_hda_codec

There was another module being loaded "snd_hda_codec_hdmi" which I blacklisted (in /etc/modprobe.d/), in hope it would stop the HDMI cards from being loaded.

After rebooting I got:

%  aplay -l | grep card
card 0: PCH [HDA Intel PCH], device 0: 92HD91BXX Analog [92HD91BXX Analog]
card 2: HDMI_1 [HDA Intel HDMI], device 3: ID 2807 Digital [ID 2807 Digital]

So I was able to stop 2 of the HDMI devices, but one still shows up... I try suspending and waking the computer, the problem persists and a new audio device appears (for card 1 again):

%  aplay -l | grep card
card 0: PCH [HDA Intel PCH], device 0: 92HD91BXX Analog [92HD91BXX Analog]
card 1: HDMI [HDA ATI HDMI], device 3: ID aa01 Digital [ID aa01 Digital]
card 2: HDMI_1 [HDA Intel HDMI], device 3: ID 2807 Digital [ID 2807 Digital]

So I'm thinking the problem must be related to these card 1 devices, but to be honest I'm happy just loading the modules for card 0 as I don't use the other devices.

Can I get any input on this?

Offline

Board footer

Powered by FluxBB