Re: [Jack-Devel] problem with jack's watchdog killing jackd process after a 3-4 seconds of operation

Yup with -P or -C alone, jack runs. I could call this discussion 'sensibly
closed',
but I'd like to check a few more things with those of you who have
successfully run jack in a  vm guest.

My virtualbox host, is a server (slackware too). so no rt kernel, no
low-latency desktop and the rest. could that affect the ability of the alsa
on the host to provide duplex audio functionality?
In virtualbox I am using the 'alsa' driver and exposing the AC97 codec on
the guest. HDA intel didn't offer much different results though.

Thank you for your help!
unfortunately, the plan to have this vm guest as a fx engine can't be
utilized :(

BUT (just had a -hopefully- brilliant idea) what if I forget about alsa,
and I use this jack server as a slave using the net backend????
could i do that and forget about the real hw ports?



On Fri, May 11, 2012 at 4:05 PM, Jonathan Woithe <[hidden]> wrote:

> HI Athanasios
>
> On Fri, May 11, 2012 at 03:28:40PM +0300, Athanasios Silis wrote:
> > i'm trying to setup jack in a slackware box.
> >
> > I have installed and enabled set_rlimits and can get jack to powerup ...
> > :
> > Then the watchdog kills the jack process. ...
> > :
> > I should inform you that i have setup this slackware as a virtualbox
> guest.
>
> For what it's worth, I successfully use set_rlimits myself on a Slackware
> box and have been doing so for many years.  Therefore I doubt we're dealing
> with a fundamental problem in the audio stack (set_rlimits, jack or ALSA).
> As Paul said, it's more likely something to do with the way the audio
> hardware is being virtualised in your VM - given real hardware the software
> mix that you're using is fine.
>
> Regards
>   jonathan
>

1336747398.15038_0.ltw:2,a <CADgchnt078P9Hz9eho9oyfhO=11VbDVCFH=LWBRPTjfQnb7-CA at mail dot gmail dot com>