Colin Guthrie wrote: > The other way round seems to work fine, e.g. starting playing with a > real sink, and removing the said sink, a null-sink is automatically > loaded and the recover-streams kicks in and works. I think I'll revise that and say the other way round is *better* but I have now had a couple failures this way round too with the same symptoms... mplayer freezes but other connections can happily continue and play sound. I'm attaching my example config and a load.sh and unload.sh scripts to load the alsa sink and unload it again. Using this setup it's hopefully quite easy to replicate this freeze which may actually highlight an underlying problem elsewhere.....? HTHs Col -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: load.sh URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20080312/ce86c58c/attachment.asc> -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: unload.sh URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20080312/ce86c58c/attachment.txt> -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: test.pa URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20080312/ce86c58c/attachment-0001.asc>
