[opensuse] Remote audio problem
Hi, I'm setting up a media machine, currently using oS12.3. It is attached to my TV as a monitor and the Hi-Fi for audio output. When I log in directly, all works as expected. When I log in remotely with ssh -X user@host from my main machine however, I can run my chosen audio player (clementine) but when trying to play any audio I get a "Connection refused" error unless the user is also logged in on the media box. This rather defeats the object of running the audio player remotely. How can I make the audio server (pulse) available with no logged in users on the media server (only the ssh -X connection)? Cheers Dylan -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 13/12/13 09:06, Dylan wrote:
Hi,
I'm setting up a media machine, currently using oS12.3. It is attached to my TV as a monitor and the Hi-Fi for audio output. When I log in directly, all works as expected.
When I log in remotely with ssh -X user@host from my main machine however, I can run my chosen audio player (clementine) but when trying to play any audio I get a "Connection refused" error unless the user is also logged in on the media box. This rather defeats the object of running the audio player remotely. How can I make the audio server (pulse) available with no logged in users on the media server (only the ssh -X connection)?
Cheers Dylan
Dylan, Take a look at Music Player Daemon <http://www.musicpd.org/>. I have it installed on my main machine here, and the daemon starts automatically after boot-up. It can be controlled from the command line, but there are also several graphical interfaces, my preference being GMPC. I can control the audio output from my laptop (also using GMPC) or my Android phone, using Droid MPD client. MPD and GMPC are available in Packman. Bob - -- Bob Williams System: Linux 3.7.10-1.16-desktop Distro: openSUSE 12.3 (x86_64) with KDE Development Platform: 4.11.4 Uptime: 06:00am up 13:53, 3 users, load average: 0.05, 0.16, 0.23 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlKq1PwACgkQ0Sr7eZJrmU6UnACfUUTlj7vRkMDcNzb0PwtbB9bq cpQAoJ/8FtL+DmoCtJXVk7njALq9kL7u =qX/h -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Friday 13 of December 2013 09:06:28 Dylan wrote:
I'm setting up a media machine, currently using oS12.3. It is attached to my TV as a monitor and the Hi-Fi for audio output. When I log in directly, all works as expected.
When I log in remotely with ssh -X user@host from my main machine however, I can run my chosen audio player (clementine) but when trying to play any audio I get a "Connection refused" error unless the user is also logged in on the media box. This rather defeats the object of running the audio player remotely. How can I make the audio server (pulse) available with no logged in users on the media server (only the ssh -X connection)?
Try setting load-module module-native-protocol-tcp auth-ip-acl=[IP ADDRESS OF CLIENT] in /etc/pulse/system.pa, then pulseaudio -D as root. This will make pulseaudio act as a server on TCP port 4713, which you'll need to open in the firewall configuration. Then in the client, PULSE_SERVER=[IP ADDRESS OF SERVER] clementine will connect to the pulseaudio daemon on the server. These directions provide a minimum of security, because the connection is authenticated by IP address. You could also configure a ssh tunnel for this or sync the pulseaudio cookie between server and client. There is enough documentation about authentication methods regarding pulseaudio on the internet.
Cheers Dylan
Regards, Peter -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 13/12/13 09:53, auxsvr@gmail.com wrote:
On Friday 13 of December 2013 09:06:28 Dylan wrote:
I'm setting up a media machine, currently using oS12.3. It is attached to my TV as a monitor and the Hi-Fi for audio output. When I log in directly, all works as expected.
When I log in remotely with ssh -X user@host from my main machine however, I can run my chosen audio player (clementine) but when trying to play any audio I get a "Connection refused" error unless the user is also logged in on the media box. This rather defeats the object of running the audio player remotely. How can I make the audio server (pulse) available with no logged in users on the media server (only the ssh -X connection)?
Try setting
load-module module-native-protocol-tcp auth-ip-acl=[IP ADDRESS OF CLIENT]
in /etc/pulse/system.pa, then
pulseaudio -D
as root. This will make pulseaudio act as a server on TCP port 4713, which you'll need to open in the firewall configuration. Then in the client,
PULSE_SERVER=[IP ADDRESS OF SERVER] clementine
will connect to the pulseaudio daemon on the server.
These directions provide a minimum of security, because the connection is authenticated by IP address. You could also configure a ssh tunnel for this or sync the pulseaudio cookie between server and client. There is enough documentation about authentication methods regarding pulseaudio on the internet.
I'm not trying to configure pulseaudio as a network audio server... I'm trying to get an application running on the media box via ssh to connect to its *local* pulse server even though there is no local user logged in. I have set pulseaudio to run in system mode but still no audio... -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box? It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station. -- _____________________________________ ---This space for rent--- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback. Dx
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/13/2013 12:52 PM, Dylan wrote:
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback.
Dx
Well, again, go back and see if your ssh connection is forwarding the proper port for the audio. Check this site:
http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Netwo...
Or more tersely: http://superuser.com/questions/231920/forwarding-audio-like-x-in-ssh -- Explain again the part about rm -rf / -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 13/12/13 21:13, John Andersen wrote:
On 12/13/2013 12:52 PM, Dylan wrote:
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback.
Dx
Well, again, go back and see if your ssh connection is forwarding the proper port for the audio.
Check this site:
http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Netwo...
Or more tersely: http://superuser.com/questions/231920/forwarding-audio-like-x-in-ssh
I really don't see why that should be relevant. I use: ssh -X dylan@zen when I run clementine in that login, it executes *on zen* (the media 'server') and its X display is forwarded to the client. Any audio connection it is trying to make is local to zen. The problem is that pulseaudio does not get run until a user logs in. Once I log in *locally* the instance of clementine running through ssh can play audio correctly. My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login. Dx -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/13/2013 1:34 PM, Dylan wrote:
On 13/12/13 21:13, John Andersen wrote:
On 12/13/2013 12:52 PM, Dylan wrote:
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback.
Dx
Well, again, go back and see if your ssh connection is forwarding the proper port for the audio.
Check this site:
http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Netwo...
Or more tersely: http://superuser.com/questions/231920/forwarding-audio-like-x-in-ssh
I really don't see why that should be relevant. I use:
ssh -X dylan@zen
when I run clementine in that login, it executes *on zen* (the media 'server') and its X display is forwarded to the client. Any audio connection it is trying to make is local to zen. The problem is that pulseaudio does not get run until a user logs in. Once I log in *locally* the instance of clementine running through ssh can play audio correctly. My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
Dx
Ok. Have it your way. Horse -->| water. -- _____________________________________ ---This space for rent--- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 13 December 2013 21:44, John Andersen <jsamyth@gmail.com> wrote:
On 12/13/2013 1:34 PM, Dylan wrote:
On 13/12/13 21:13, John Andersen wrote:
On 12/13/2013 12:52 PM, Dylan wrote:
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote:
I get a "Connection refused" error unless the user is also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback.
Dx
Well, again, go back and see if your ssh connection is forwarding the proper port for the audio.
Check this site:
http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Netwo...
Or more tersely: http://superuser.com/questions/231920/forwarding-audio-like-x-in-ssh
I really don't see why that should be relevant. I use:
ssh -X dylan@zen
when I run clementine in that login, it executes *on zen* (the media 'server') and its X display is forwarded to the client. Any audio connection it is trying to make is local to zen. The problem is that pulseaudio does not get run until a user logs in. Once I log in *locally* the instance of clementine running through ssh can play audio correctly. My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
Dx
Ok. Have it your way. Horse -->| water.
You have completely misunderstood the question, refused to listen to explanations of how you are mistaken, and are now making yourself look like an idiot by being obnoxious about it. Good job. OP: Why bother with PA at all? Surely it would be easier just to use alsa directly? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 13/12/13 21:58, Aneurin Price wrote:
On 13 December 2013 21:44, John Andersen <jsamyth@gmail.com> wrote:
On 12/13/2013 1:34 PM, Dylan wrote:
On 13/12/13 21:13, John Andersen wrote:
On 12/13/2013 12:52 PM, Dylan wrote:
On 13/12/13 19:32, John Andersen wrote:
On 12/13/2013 2:33 AM, Dylan wrote: > I get a "Connection refused" error unless the user is > also logged in on the media box.
Where does this message appear? SSH station, or media box?
It might be your ssh station that is refusing connection because you didn't allow port forwarding on either the ssh server (media box) or the ssh station.
It is a message from the instance of clementine running via ssh - it displays and functions correctly displayed on the client. The only thing which doesn't work as expected is audio playback.
Dx
Well, again, go back and see if your ssh connection is forwarding the proper port for the audio.
Check this site:
http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Netwo...
Or more tersely: http://superuser.com/questions/231920/forwarding-audio-like-x-in-ssh
I really don't see why that should be relevant. I use:
ssh -X dylan@zen
when I run clementine in that login, it executes *on zen* (the media 'server') and its X display is forwarded to the client. Any audio connection it is trying to make is local to zen. The problem is that pulseaudio does not get run until a user logs in. Once I log in *locally* the instance of clementine running through ssh can play audio correctly. My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
Dx
Ok. Have it your way. Horse -->| water.
You have completely misunderstood the question, refused to listen to explanations of how you are mistaken, and are now making yourself look like an idiot by being obnoxious about it. Good job.
Thank you - I was beginning to think I had fundamentally misunderstood ssh all these years...
OP: Why bother with PA at all? Surely it would be easier just to use alsa directly?
I want to be able to balance audio from various sources and pass them to different outputs without all the fiddling around with jack which is massively over-specified for my needs. PA does that fine for me - I have the motherboard audio device, a soundcard, and a usb-audio mixer (each connected to a separate amp) which together with PA allows me to route audio from software or external input(s) to various parts of the flat dynamically - but currently only from the server itself. Dx
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 14 December 2013 00:19, Dylan <dylan@dylan.me.uk> wrote:
OP: Why bother with PA at all? Surely it would be easier just to use alsa directly?
I want to be able to balance audio from various sources and pass them to different outputs without all the fiddling around with jack which is massively over-specified for my needs. PA does that fine for me - I have the motherboard audio device, a soundcard, and a usb-audio mixer (each connected to a separate amp) which together with PA allows me to route audio from software or external input(s) to various parts of the flat dynamically - but currently only from the server itself.
Interesting. I can't help with PA specifically as I generally find it causes more problems than it solves, but I'd probably start by looking at what gets run in a full session - any PA related processes, any relevant environment variables, possibly dbus, including all arguments. Then I'd try putting all those into a script and running that in my ssh command rather than just running Clementine as a single command. If that doesn't prove fruitful, I'd seriously consider the possibility of changing the approach to the problem. For example, run a full desktop session via xvnc and keep that running persistently, then connect to that in place of 'ssh -X user@host clementine'. (I can be more specific about that if you do end up going in this direction.) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 14/12/13 12:17, Aneurin Price wrote:
On 14 December 2013 00:19, Dylan <dylan@dylan.me.uk> wrote:
OP: Why bother with PA at all? Surely it would be easier just to use alsa directly?
I want to be able to balance audio from various sources and pass them to different outputs without all the fiddling around with jack which is massively over-specified for my needs. PA does that fine for me - I have the motherboard audio device, a soundcard, and a usb-audio mixer (each connected to a separate amp) which together with PA allows me to route audio from software or external input(s) to various parts of the flat dynamically - but currently only from the server itself.
Interesting. I can't help with PA specifically as I generally find it causes more problems than it solves, but I'd probably start by looking at what gets run in a full session - any PA related processes, any relevant environment variables, possibly dbus, including all arguments. Then I'd try putting all those into a script and running that in my ssh command rather than just running Clementine as a single command.
I hadn't thought about that - will have a look at that. Should be able to have a script run at login to check if it's running and start it if not.
If that doesn't prove fruitful, I'd seriously consider the possibility of changing the approach to the problem. For example, run a full desktop session via xvnc and keep that running persistently, then connect to that in place of 'ssh -X user@host clementine'. (I can be more specific about that if you do end up going in this direction.)
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 12/13/2013 03:34 PM, Dylan wrote:
I really don't see why that should be relevant. I use:
ssh -X dylan@zen
when I run clementine in that login, it executes *on zen* (the media 'server') and its X display is forwarded to the client. Any audio connection it is trying to make is local to zen. The problem is that pulseaudio does not get run until a user logs in. Once I log in *locally* the instance of clementine running through ssh can play audio correctly. My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
Dx
Dylan, I think what you should try is logging onto zen. Find out exactly what you need to do to start clementine for the command line. If you are starting it from a gui, then open an xterm and look at ps axf for the default options clementine is started with. If for some reason 'ssh dylan@zen /usr/bin/clementine' isn't starting it the way you want, you can add the options to your ssh dylan@zen /usr/bin/clementine [options]' -- or -- you can then create a simple bash script that starts clementine with the proper options (clementine-run.sh) and save that to your home/scr[ipt] or /home/bin or whereever. (make sure it is executable) Like: #!/bin/bash /usr/bin/clementine -a blah -d blah -p bla ssh has the ability to execute whatever you want from your remote box on zen with the form: ssh user@hostname command ssh dylan@zen /path/to/clementine-run.sh Use this to your advantage to launch clementine on zen from wherever you want. If you are not opening an X window of clementine on your remote host when you ssh -X into zen, then forget the -X, you are just starting the app on zen. If you are getting a gui back, then keep it. If clementine is bombing when you try and start it, the see if you can find some error in /var/log/message or in ~/.xsession-errors If I have totally missed what you are trying to do, then I apologize. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Fri, 13 Dec 2013 21:34:49 +0000 Dylan <dylan@dylan.me.uk> wrote:
My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
In 13.1 pulseaudio is set NOT to start as system process by default. To change that set PULSEAUDIO_SYSTEM="no" to "yes" in /etc/sysconfig/sound . For me next would be reboot. Some systemd expert may come up with idea how to start audio without rebooting. The other option is to use: YaST > System > "/etc/sysconfig Editor" then in a left panel Hardware > Sound > PulseAudio > Systemwide change PULSEAUDIO_SYSTEM to "yes". YaST may restart sound system without rebooting. -- Regards, Rajko. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 15/12/13 00:48, Rajko wrote:
On Fri, 13 Dec 2013 21:34:49 +0000 Dylan <dylan@dylan.me.uk> wrote:
My problem is to get pulseaudio to run system-wide at boot, rather than when a user logs in locally to a graphical login.
In 13.1 pulseaudio is set NOT to start as system process by default. To change that set PULSEAUDIO_SYSTEM="no" to "yes" in /etc/sysconfig/sound .
Thanks, but that's not available in the 12.3 system I'm working on... PA is also set NOT to start as a system process there, but the setting is not available in sysconfig... Dx -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (7)
-
Aneurin Price -
auxsvr＠gmail.com -
Bob Williams -
David C. Rankin -
Dylan -
John Andersen -
Rajko