From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugtrack@alsa-project.org Subject: [ALSA - lib 0001184]: xmms/beep "couldn't open audio" on song change Date: Mon, 15 Aug 2005 14:59:27 +0200 Message-ID: <9a85332c7b13d2d64d3fe6a4f0fb3660@bugtrack.alsa-project.org> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from bugtrack.alsa-project.org (unknown [195.113.161.147]) by alsa.jcu.cz (ALSA's E-mail Delivery System) with ESMTP id B9D7513D for ; Mon, 15 Aug 2005 14:59:27 +0200 (MEST) Sender: alsa-devel-admin@lists.sourceforge.net Errors-To: alsa-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: alsa-devel@alsa-project.org List-Id: alsa-devel@alsa-project.org The following issue has been ASSIGNED. ====================================================================== ====================================================================== Reported By: Xero Assigned To: perex ====================================================================== Project: ALSA - lib Issue ID: 1184 Category: 0_general Reproducibility: always Severity: major Priority: normal Status: assigned ====================================================================== Date Submitted: 06-11-2005 21:14 CEST Last Modified: 08-15-2005 14:59 CEST ====================================================================== Summary: xmms/beep "couldn't open audio" on song change Description: When playing a song in beep sometimes it'll pop up a "couldn't open audio" box as if the device is in use. This is using dmix, however, so that seemed a bit weird. I normally have to close and reopen beep to resolve this. I believe xmms does the same thing because I've read about people having the exact same issue with xmms. This starts showing up when I start pressing play again: ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: Connection refused ALSA lib pcm_dmix.c:834:(snd_pcm_dmix_open) unable to connect client ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: No such file or directory ALSA lib pcm_dmix.c:834:(snd_pcm_dmix_open) unable to connect client ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: No such file or directory ALSA lib pcm_dmix.c:834:(snd_pcm_dmix_open) unable to connect client ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: No such file or directory ALSA lib pcm_dmix.c:834:(snd_pcm_dmix_open) unable to connect client ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: No such file or directory However one thing which I've never seen mentioned in the previous reports, is that when trying to use another app through dmix it doesn't work either. Here's what mplayer says, when trying to play a video: alsa-init: 1 soundcard found, using: default ALSA lib pcm_direct.c:196:(make_local_socket) connect failed: /tmp/alsa-dmix-30443-1118514847-814284: No such file or directory ALSA lib pcm_dmix.c:834:(snd_pcm_dmix_open) unable to connect client alsa-init: playback open error: No such file or directory Could not open/initialize audio device -> no sound. Audio: no sound so apparently something went wrong with /tmp/alsa-dmix-*. As soon as I closed beep, all other applications started working properly again, and when I start it back up again, it works fine again too, until it does this again. It seems to do it sporadically so it's somewhat hard to test for. Attached is my asoundrc. ====================================================================== ---------------------------------------------------------------------- Xero - 06-11-05 21:14 ---------------------------------------------------------------------- Oh, I forgot to mention, my card is an echo mia midi...if that matters. ---------------------------------------------------------------------- tiwai - 06-15-05 15:03 ---------------------------------------------------------------------- Try the CVS version. This might be fixed because of the recent change of the default dmix config option. ---------------------------------------------------------------------- Xero - 06-15-05 22:31 ---------------------------------------------------------------------- I made a CVS alsa-lib ebuild and am now running cvs alsa-lib with the ipc_gid audio option in the dmix.conf. However, since my local config file overrides the default dmix config with !dmix I've added it to my .asoundrc as well just in case. I will report back whether or not the problem still occurs. It happens somewhat randomly, so it's not the easiest thing to test for. ---------------------------------------------------------------------- Xero - 06-15-05 23:55 ---------------------------------------------------------------------- ALSA lib pcm_direct.c:218:(make_local_socket) connect failed: /tmp/alsa-dmix-27521-1118870832-290176: Connection refused ALSA lib pcm_dmix.c:847:(snd_pcm_dmix_open) unable to connect client ALSA lib pcm_direct.c:218:(make_local_socket) connect failed: /tmp/alsa-dmix-27521-1118870832-290176: No such file or directory ALSA lib pcm_dmix.c:847:(snd_pcm_dmix_open) unable to connect client it happened again. I was just playing a track and chatting and it just popped up the cannot connect. I had to restart beep before it started working again. So the CVS change didn't do anything. ---------------------------------------------------------------------- tiwai - 06-22-05 18:22 ---------------------------------------------------------------------- Weird... Could you check whether these socket files still exist if this happens? Also, check which process opens /dev/snd/pcmC0D0p (e.g. via fuser or lsof). ---------------------------------------------------------------------- Xero - 06-22-05 19:01 ---------------------------------------------------------------------- I know for a fact the socket still exists after this happens, infact it usually ends up leaving a bunch of sockets because when it does happen they don't seem to get deleted. I will check which process opens it as soon as I get a chance, probably later today. ---------------------------------------------------------------------- Xero - 06-23-05 02:47 ---------------------------------------------------------------------- here's lsof output # lsof /dev/snd/pcmC0D0p COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME beep-medi 5075 xero mem CHR 116,16 5808 /dev/snd/pcmC0D0p beep-medi 5075 xero 27u CHR 116,16 5808 /dev/snd/pcmC0D0p beep-medi 32106 xero mem CHR 116,16 5808 /dev/snd/pcmC0D0p beep-medi 32106 xero 27u CHR 116,16 5808 /dev/snd/pcmC0D0p ---------------------------------------------------------------------- Xero - 06-26-05 22:08 ---------------------------------------------------------------------- I think this tends to happen more under load. I'm starting to wonder if this is relevant to the sound card driver. I also get distortion with dmix sometimes as well when under load, pops, clicks, and occasionally the fade in of metallic sounding distortion, I've posted in bug 0000881 about my problems with that, though at this point I'm thinking these two things are relevant to each other. This has become extremely annoying and I'm not too sure what to do. Even when I don't use dmix beep will just tend to freeze completely every once in a while, usually when under load as well. ---------------------------------------------------------------------- tiwai - 06-27-05 16:22 ---------------------------------------------------------------------- The noises of dmix on a heavy CPU load is considerable, but this should have nothing to do with the busy-state of socket. The confliction of socket means a different thing, e.g. the root takes the socket and sets it as the permission 0600. No certain clue yet, though. ---------------------------------------------------------------------- Xero - 06-27-05 22:05 ---------------------------------------------------------------------- no, the socket remains the proper permission and owned by my user as well, usually. I don't really ever run anything as root, especially audio stuff, so I have no idea how that'd even happen. Let me wait for it to happen again and I'll post a ls -l of the socket. All I have to do is wait for it to happen a few times and there will be quite the collection of dead sockts in there... ---------------------------------------------------------------------- Xero - 06-27-05 22:41 ---------------------------------------------------------------------- hrm the socket actually did disappear this time when it happened, weird. That means it may or may not always remove the socket, I'm curious if the socket gets removed after i press play again when it first happens vs. just shutting down beep right then. I'll report back on this later when I have more of a clue. This is somewhat of a pain to reproduce as it only really happens on song change...and then only sometimes. ---------------------------------------------------------------------- skandalfo - 08-01-05 23:43 ---------------------------------------------------------------------- This seems to be hitting xine-lib too. See: https://sourceforge.net/tracker/index.php?func=detail&aid=1226595&group_id=9655&atid=109655 ---------------------------------------------------------------------- Xero - 08-01-05 23:50 ---------------------------------------------------------------------- I only notice this when using dmix...which I've not been using recently seeing as this card has 4 virtual stereo channels and all these problems got on my neves...it's enough to avoid having to use it...however I still get random distortion every once in a while...I'm thinking that distortion and this problem are caused by some sort of issue related to the card and dma inaccuracies...though I'm really not sure... Issue History Date Modified Username Field Change ====================================================================== 06-11-05 21:14 Xero New Issue 06-11-05 21:14 Xero File Added: asoundrc 06-11-05 21:14 Xero Note Added: 0004994 06-15-05 15:03 tiwai Note Added: 0005030 06-15-05 22:31 Xero Note Added: 0005046 06-15-05 23:55 Xero Note Added: 0005049 06-22-05 18:22 tiwai Note Added: 0005142 06-22-05 19:01 Xero Note Added: 0005144 06-23-05 02:47 Xero Note Added: 0005162 06-26-05 22:08 Xero Note Added: 0005267 06-27-05 16:22 tiwai Note Added: 0005275 06-27-05 22:05 Xero Note Added: 0005288 06-27-05 22:41 Xero Note Added: 0005289 08-01-05 23:43 skandalfo Note Added: 0005597 08-01-05 23:50 Xero Note Added: 0005598 08-15-05 14:59 perex Status new => assigned 08-15-05 14:59 perex Assigned To => perex ====================================================================== ------------------------------------------------------- SF.Net email is Sponsored by the Better Software Conference & EXPO September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf