Enrique, Your information has been most helpful. Unfortunately, the system is not quite working.
The 1 second sleep didn't quite do it. I added a 5 second sleep after starting nasd and esd. This caused them to remain running in the background. I have the following lines in my lts.conf to enable the sound: SOUND = Y SOUND_DAEMON = nasd SMODULE_01 = cmpci.o The server has a 1.8 GHz P4 and an integrated i810 sound module while the workstation has a 1.0 GHz Celeron and an integrated CM8738 sound module. I noticed that the rc.sound that comes with the RPMs does not include an "insmod soundcore.o" as the file that came with the ltsp tar sound package. I added a line to load soundcore.o. It did load but it didn't seem to make any difference. BTW, I am using some OSS sound package on my server and it doesn't use soundcore but does load a bunch of stuff in "modules.conf". I have nothing in my workstation modules.conf. Do I need anything in the workstation "modules.conf"? If I log into the workstation as a different user than the server, I get error messages indicating that /dev/dsp cannot be accessed due to a permissions issue. If I log in as the same user, it accesses ok but the only allow sound to come out of the server's speakers. I see a line while the workstation is booting that indicates "accepting connections on port 16001". The client has an ip address of 192.168.1.102. I can run /sbin/auinfo on the workstation. It shows that the audioserver is on tcp/192.168.1.102:8000. if I type "export" on the client, from a bash shell, I see the following: ESPEAKER:192.168.1.102:16001 AUDIOSERVER:192.168.1.102:0.0 I can type "esdplay x.wav" on the workstation and the sound comes from the proper speakers. If I type "auplay x.wav", it just hangs. At that point, I can not longer type "auinfo" until I reboot the workstation. Any additional information will be greatly appreciated. Thanks, Rob ----- Original Message ----- From: "Enrique Rodriguez" <[EMAIL PROTECTED]> To: "G. Major" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Sunday, December 30, 2001 6:49 PM Subject: RE: [Ltsp-discuss] ltsp sound on 3.0 Hi, Rob, Two areas with the sound package I have seen issues with were getting the ESD and NASD services running on the LTSP client and making sure the sound modules were loading properly on the client. As you mention, a ps -ax doesn't show ESD or NASD running on the client. I checked notes and found that I had a faster processor LTSP client (PIII 500) that didn't load these services. They didn't appear to connect during boot script (rc.local calls rc.sound) execution. The solution was to "SLEEP 1" after NASD and ESD startup in /opt/ltsp/i386/etc/rc.sound. As for getting modules to load, you must enable sound and the necessary modules in lts.conf. I had the benefit of same server and client hardware to figure out which modules. Keep in mind that the order the modules are loaded matters. I used lsmod to determine which modules were in use on the server and which ones were and weren't loading on the client. I don't recall if lsmod is available on the client but it is easy enough to put it there. lsmod showed me on an i810-based LTSP client that the sound.o driver being loaded by rc.sound was "blocking" the i810 specific driver I had put in lts.conf. I had to disable the sound.o line in rc.sound for specific (i810) clients. Unrelated to your problem but of note was that the default sound volume is quite high. I set in rc.sound "aumix-minimal -v65" to prevent the clients from coming up too loud. -enrique -----Original Message----- From: G. Major [mailto:[EMAIL PROTECTED]] Sent: Saturday, December 29, 2001 9:18 PM To: [EMAIL PROTECTED]; Enrique Rodriguez Subject: Re: [Ltsp-discuss] ltsp sound on 3.0 I downloaded the rpm sound package from K12LTSP. I had previously installed the ltsp core, kernel & X packages from tar files that I obtained from ltsp.org. The sound rpm complained that the core was not installed, even though it was by tar. I used the "nodeps" flag & installed it (over the previously installed tar sound package from ltsp). The only way that I can get sound out of the workstation is by executing "auplay" to play a ".wav" file. Otherwise, I just get an error message that the system can not load the /dev/dsp when a different user is logged into the server. I do not know if it was properly initialized. When I boot the workstation, I do not see nasd, esd or anything else useful running in the background (ps -ax). There also doesn't seem to be any documentation with this. Also, does anyone know if it is alright to completely reinstall ltsp from rpms over the previously installed tar files? I am concerned that this will mess up the many various scripts that have been modified. Does anyone have any technical information and also a way to test the installation. Thanks, Rob ----- Original Message ----- From: "Enrique Rodriguez" <[EMAIL PROTECTED]> To: "G. Major" <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]> Sent: Saturday, December 29, 2001 9:51 AM Subject: RE: [Ltsp-discuss] ltsp sound on 3.0 Hi, Rob, I wasn't sure whether you are using the ltsp sound package from the LTSP contrib area or from the K12LTSP group. Using RedHat 7.2 and LTSP 3.0 I installed the sound package provided by the K12LTSP group. I did not modify any install.sh. It comes as an RPM so I installed it the RPM way. You might want to try (if you aren't already): ltsp_sound-0.1c-8.i386.rpm at: ftp://k12linux.mesd.k12.or.us/pub/distributions/ISO/K12LTSP-7.2/RPMS/ -enrique -----Original Message----- From: G. Major [mailto:[EMAIL PROTECTED]] Sent: Friday, December 28, 2001 7:05 PM To: [EMAIL PROTECTED] Subject: [Ltsp-discuss] ltsp sound on 3.0 I have been attempting to get sound working with ltsp 3.0 for some time. I have ltspsound-0.1c. I am running RH7.2 on the server. The server has an integrated i810 sound module. The workstation has an integrated cm8738 device. I modified it to deal with the ltsp 3.0 structures. I manually created the /opt/ltsp/i386/dev, dsp1 & mixer entries, based upon the "install.sh" script. I modified the ltsp.conf file to load "nasd" as SOUND_DAEMON and "cmpci.o" as SMODULE_01. If the same user is logged into the server and workstation, the system boots and sound comes out of the server's speakers. If different users are logged in, I get an error message indicating that it cannot load the /dev/dsp. The owner of the /dev/dsp0,1,2,3 & 56k plus the four mixers (plus link) on the server, is changed to whomever is logged in while the group is root. I can run "auplay" or "play" with WAV files on the workstation and hear the sound on the workstation's speakers. The default sounds do not come out of these speakers. My ultimate goal is to have OSS sound support working on the workstation. Can anyone please offer any suggestions about getting this to work properly? Thanks, Rob _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com _____________________________________________________________________ Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto: https://lists.sourceforge.net/lists/listinfo/ltsp-discuss For additional LTSP help, try #ltsp channel on irc.openprojects.net _____________________________________________________________________ Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto: https://lists.sourceforge.net/lists/listinfo/ltsp-discuss For additional LTSP help, try #ltsp channel on irc.openprojects.net
