Program is crashing at least once an hour for me. May or may not be related to my use of gkrellmms, but either way, this wasn't happening with previous versions. More information available upon request. Reproducible: Sometimes Steps to Reproduce:
I've been running it quite stable with almost every xmms related plugin in portage installed... so I'm going to need more information from you. Can you please compile it with debugging information turned on? To do this, you might want to set something like this up at the end of your /etc/make.conf file and uncommern it. Note that the -g is the important part of the FLAGS # # Debug options # CFLAGS="-march=pentium3 -mcpu=pentium3 -pipe -g" CXXFLAGS="${CFLAGS}" USE="${USE} debug" FEATURES="${FEATURES} nostrip" Then please run it through gdb and type 'bt' to produce a backtrace when xmms crashes: $ gdb xmms <blah> (gdb) run <blah> SEGFAULT (gdb) bt <blah> copy to this bug report the important (tail) section of the gdb output starting just before the segfault.
Following error is ... a bit odd, so far as I'm concerned. Versions: linux kernel 2.6.1, glib-1.2.10-r5, glibc-2.3.3_pre20040117. GLib-ERROR **: could not allocate 1076164900 bytes aborting... Program received signal SIGABRT, Aborted. [Switching to Thread 229381 (LWP 23247)] 0x40488711 in kill () from /lib/libc.so.6 (gdb) bt #0 0x40488711 in kill () from /lib/libc.so.6 #1 0x40416a60 in pthread_kill () from /lib/libpthread.so.0 #2 0x40416dbb in raise () from /lib/libpthread.so.0 #3 0x404884a2 in raise () from /lib/libc.so.6 #4 0x40489a9c in abort () from /lib/libc.so.6 #5 0x4023f246 in g_logv () from /usr/lib/libglib-1.2.so.0 #6 0x4023f284 in g_log () from /usr/lib/libglib-1.2.so.0 #7 0x4023dc50 in g_malloc0 () from /usr/lib/libglib-1.2.so.0 #8 0x00000004 in ?? () #9 0x4024c86a in days_in_year () from /usr/lib/libglib-1.2.so.0 #10 0x4024f924 in __JCR_LIST__ () from /usr/lib/libglib-1.2.so.0 #11 0x0000f6c0 in ?? () #12 0x0006da6d in ?? () #13 0x00000008 in ?? () #14 0x40098ad0 in __JCR_LIST__ () from /usr/lib/libxmms.so.1
Okay... so an error in allocating ~1GB of something... seems kinda large, if that's what you were referring to being "odd". Immediately I notice your glibc version... very new glib appears to be the source of the problem though, i'd rebuild glib... it's small, it should go quickly. emerge info output might be helpful in solving this...
We're ignoring the fact that xmms-1.2.8-r3 works fine? (I moved back to it, after capturing the above).
See this bug for some relevant looking information: http://bugs.xmms.org/show_bug.cgi?id=1583 Also, xmms 1.2.9 was released today.
I have uploaded the gentoo patchset to be mirrored... I will have xmms-1.2.9 added to portage sometime tomorrow when I see the patchset has been properly mirrored.
Same bug is going to be in 1.2.9, going by developer's response over at bugs.xmms.org
For now, disable the rva and imms plugins (suggested on the xmms bug). I'll watch that bug and merge in the fix when it's made available. If you haven't already done so, you might want to provide some more info on the xmms bug
There's a fix mentioned today in the xmms bug. Can someone with this problem try out that fix to see if it works. If it does, I'll include it in the next xmms patchset.
Ok... I've tried out the fix mentioned on the xmms bugzilla, and I've had xmms running with imms and rva for a while now with no notice of a problem... so I've committed it in 1.2.9-r2.
I'm having the same problem, although it's not exactly reproducible. It seems to happen whenever I'm really particularly enjoying a song, but since I don't enjoy a song as much the second time round, it's difficult to reproduce it twice in a row. The fixes mentioned above seem to have worked, though - I've played some pretty good songs, without any crashes. I can't say for sure, yet, though. One thing, though: I am also getting frequent errors from Gkrellm. That is, it exits (it terminates itself, it doesn't crash), and gives me a friendly little xmessage box: xmessage gkrellm aborted: GkrellMMS (update_monitor) I'm not sure what this means, but it doesn't seem linked to anything, really. It started happening at about the same time as xmms 1.2.9 started crashing, though, which is why I mention it in this bug. I've tried gdb'ing gkrellm, but it's not easy, since gkrellm doesn't crash - it seems perfectly happy with the situation, simply exiting gracefully with the above message.
Okay, it is the same problem: GLib-ERROR **: gmem.c:157: failed to allocate 1083676732 bytes aborting... gkrellm aborted: GkrellMMS (update_monitor) Aborted
Jonathan. Please open a new bug for the gkrellm problem
Reopening as xmms just committed an official patch for the problem that I'd prefer to use instead of our fix in the next patchset.
this is fixed in 1.2.10, and an ebuild will be up soon...
