The process in question is jackd, and once this stuck state is reached one cannot even start a new instance of jack which leads me to believe this is a jack issue, particularly considering this is not a problem on Ubuntu 12.04 where you could still kill the process under these circumstances, even though I still don't think allowing jack to get stuck like this is the best approach to dealing with an issue like this, however unlikely it may be. Now, it may be a case of bad packaging  or some out-of-date version, but given the circumstances I'm not convinced this is an alsa issue at this point in time.

-- 
Ivica Ico Bukvic, D.M.A.
Associate Professor
Computer Music
ICAT Senior Fellow
Director -- DISIS, L2Ork
Virginia Tech
School of Performing Arts – 0141
Blacksburg, VA 24061
(540) 231-6139
ico@vt.edu
www.performingarts.vt.edu
disis.icat.vt.edu
l2ork.icat.vt.edu
Ico.bukvic.net

On Oct 17, 2015 2:44 PM, "Fons Adriaensen" <fons@linuxaudio.org> wrote:
On Sat, Oct 17, 2015 at 07:24:02PM +0100, Will Godfrey wrote:
> On Sat, 17 Oct 2015 13:58:37 -0400
> Ivica Ico Bukvic <ico@vt.edu> wrote:
>
> > Hi all,
> >
> > This is a long-standing problem. While it is not intentional, sometimes
> > newcomers to jack on Linux tend to pull out the soundcard (USB) before
> > shutting down jack. This results in jack permanently hanging to the
> > point where one has to force-shutdown the machine. That is at least the
> > case on Ubuntu 14.04 (and was on 12.04) with lowlatency kernel. Trying
> > to do sudo killall -9 jackd makes no difference. Essentially, it is
> > impossible to destroy the process and reboot hangs because of it.
> >
> > Any idea what can be done to minimize this problem or alleviate it
> > altogether?
> >
> > Best,
> >
> > Ico
>
>
> a workaround is from a terminal type:
>
> pidof jackd
>
> you'll get a number such as 3772
>
> then type
>
> kill -9 3772

That's what killall or pkill will do anyway, so
I don't think this is a solution.

I also don't think that Jack is involved. You'd
probably get the same effect when unplugging the
USB device while some ALSA app is using it.

There is a similar problem with the hdsp-madi module,
and it has hit me a number of times when installing
the card in a new system. Apparently the default
configuration expects an external clock, and when
that isn't connected any process that uses the device
hangs and can't be killed. The first process to do so
will be alsactl, called as part of the boot sequence.
And since that will be called again as part of reboot
or poweroff, those will hang as well. The only thing
you can do to pull the power plug, remove the card,
boot, edit the file used by alsactl, switch off, put
the card back and boot.

Ciao,

--
FA

A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user