Several times now after ripping for a while I get this crash. I get the g_warning() pretty early on though, so the bug may happen earlier and just takes a while to turn into a segfault. (sound-juicer:4114): GStreamer-CRITICAL **: file gstbin.c: line 1259 (gst_bin_iterate): assertion `GST_IS_BIN (bin)' failed (sound-juicer:4114): GStreamer-CRITICAL **: file gstbin.c: line 1259 (gst_bin_iterate): assertion `GST_IS_BIN (bin)' failed Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -157276480 (LWP 4114)] gst_bin_iterate (bin=0xa002758) at gstbin.c:1259 1259 g_return_val_if_fail (GST_IS_BIN (bin), FALSE); (gdb) bt #0 gst_bin_iterate (bin=0xa002758) at gstbin.c:1259 #1 0x00278a98 in g_idle_dispatch (source=0xa0cabd8, callback=0x4, user_data=0x9d6dda0) at gmain.c:3806 #2 0x0027574b in g_main_context_dispatch (context=0x9dead78) at gmain.c:1942 #3 0x002771d2 in g_main_context_iterate (context=0x9dead78, block=1, dispatch=1, self=0x9d67360) at gmain.c:2573 #4 0x0027747f in g_main_loop_run (loop=0x9fc8098) at gmain.c:2777 #5 0xf6ce778e in gtk_main () at gtkmain.c:1173 #6 0x08051416 in main (argc=4, argv=0x4) at sj-main.c:1085 (gdb) info locals __t = 165076384 __r = 165076384 running = 33668215 __PRETTY_FUNCTION__ = "gst_bin_iterate" (gdb) up #1 0x00278a98 in g_idle_dispatch (source=0xa0cabd8, callback=0x4, user_data=0x9d6dda0) at gmain.c:3806 3806 return callback (user_data); (gdb) info locals No locals. (gdb) list 1254 gst_bin_iterate (GstBin * bin) 1255 { 1256 gboolean running; 1257 1258 g_return_val_if_fail (bin != NULL, FALSE); 1259 g_return_val_if_fail (GST_IS_BIN (bin), FALSE); 1260 (gdb) p *bin $1 = {element = {object = {object = {g_type_instance = {g_class = 0x4}, ref_count = 1, qdata = 0xf8}, name = 0x20 <Address 0x20 out of bounds>, lock = 0xa001520, parent = 0x7, flags = 167784408, _gst_reserved = {0x0, 0x0, 0x0, 0x0}}, current_state = 0 '\0', pending_state = 0 '\0', loopfunc = 0, sched = 0x0, sched_private = 0x0, clock = 0x0, base_time = 0, numpads = 0, numsrcpads = 0, numsinkpads = 0, pads = 0x0, state_mutex = 0x0, state_cond = 0x0, pre_run_func = 0, post_run_func = 0, prop_value_queue = 0x0, property_mutex = 0x0, _gst_reserved = {0x0, 0x0, 0x0, 0x0}}, numchildren = 0, children = 0x0, child_states = { GST_STATE_VOID_PENDING, GST_STATE_VOID_PENDING, GST_STATE_VOID_PENDING, GST_STATE_VOID_PENDING}, _gst_reserved = {0x0, 0x0, 0x0, 0x0}}
I have some sense that this always happened at the beginning of a track, but I'm not sure.
I think I see how this *might* happen. I added a little patch which I'm 95% sure should be correctness-preserving, and might fix this problem. Can you try this version? Binary: http://people.redhat.com/walters/sound-juicer-0.5.14-3.i386.rpm Source: http://people.redhat.com/walters/sound-juicer-0.5.14-3.src.rpm
This has been pushed into rawhide; marking as RAWHIDE. Please reopen if you still have problem.s