virt-manager-0.5.2-2.fc8.x86_64 libvirt-0.4.0-1.fc8.x86_64 gtk-vnc-0.2.0-4.fc8.x86_64 gtk-vnc-python-0.2.0-4.fc8.x86_64 virt-manager starts up fine, but then when I try and connect to my local libvirt (on the same machine) it segfaults and dies. I installed all the debuginfo (50MB! :) The core gives me: #0 0x0000003ef16302de in remoteAuthenticate (conn=0x142a520, priv=0x142ab30, in_open=1, auth=0x0, authtype=0x0) at remote_internal.c:3350 #1 0x0000003ef1632378 in doRemoteOpen (conn=0x142a520, priv=0x142ab30, uri=0x1408f70, auth=0x0, flags=2) at remote_internal.c:748 #2 0x0000003ef1632fe2 in remoteOpen (conn=0x142a520, uri=0x1408f70, auth=0x0, flags=<value optimized out>) at remote_internal.c:854 #3 0x0000003ef160b9b7 in do_open (name=0xbac834 "qemu:///system", auth=0x0, flags=0) at libvirt.c:571 #4 0x00002aaab2905ae5 in libvirt_virConnectOpen (self=<value optimized out>, args=<value optimized out>) at libvirt-py.c:1026 #5 0x0000003a85cb5b79 in PyEval_EvalFrameEx (f=0x142a370, throwflag=<value optimized out>) at Python/ceval.c:3564 #6 0x0000003a85cb542c in PyEval_EvalFrameEx (f=0x142a030, throwflag=<value optimized out>) at Python/ceval.c:3650 #7 0x0000003a85cb6862 in PyEval_EvalCodeEx (co=0xf830a8, globals=<value optimized out>, locals=<value optimized out>, args=0x10475a8, argcount=1, kws=0x1423b90, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2831 #8 0x0000003a85c5968c in function_call (func=0xfacb18, arg=0x1047590, kw=0x1424960) at Objects/funcobject.c:517 #9 0x0000003a85c3c303 in PyObject_Call (func=0x142ab30, arg=0x3ef164fca8, kw=0x3ef162fece) at Objects/abstract.c:1860 #10 0x0000003a85cb3cac in PyEval_EvalFrameEx (f=0x1429e60, throwflag=<value optimized out>) at Python/ceval.c:3844 #11 0x0000003a85cb542c in PyEval_EvalFrameEx (f=0x1429c30, throwflag=<value optimized out>) at Python/ceval.c:3650 #12 0x0000003a85cb6862 in PyEval_EvalCodeEx (co=0x6f8c60, globals=<value optimized out>, locals=<value optimized out>, args=0x1047628, argcount=1, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at Python/ceval.c:2831 #13 0x0000003a85c596ee in function_call (func=0x7068c0, arg=0x1047610, kw=0x0) at Objects/funcobject.c:517 #14 0x0000003a85c3c303 in PyObject_Call (func=0x142ab30, arg=0x3ef164fca8, kw=0x3ef162fece) at Objects/abstract.c:1860 #15 0x0000003a85c434fd in instancemethod_call (func=<value optimized out>, arg=0x1047610, kw=0x0) at Objects/classobject.c:2497 #16 0x0000003a85c3c303 in PyObject_Call (func=0x142ab30, arg=0x3ef164fca8, kw=0x3ef162fece) at Objects/abstract.c:1860 #17 0x0000003a85caf3fe in PyEval_CallObjectWithKeywords (func=0x104fc80, arg=0x2aaaaaac4050, kw=0x0) at Python/ceval.c:3433 #18 0x0000003a85cdd31d in t_bootstrap (boot_raw=0x14299e0) at Modules/threadmodule.c:424 #19 0x0000003a73606407 in start_thread () from /lib64/libpthread-2.7.so #20 0x0000003a72ad4b0d in clone () from /lib64/libc-2.7.so Happy to test/provide more info.
Thanks for the debugging. No need for further info - the stack trace shows a clear flaw in libvirt. Will prepare a patch...
Created attachment 290591 [details] Fix crash with auth callback If you want to test a fix, the following patch applies against the libvirt RPM and ought to address the crash you discovered.
Yep. That does seem to fix it right up. ;) This can be closed as far as I am concerned... when can we expect a updated f8 package? For anyone seeing this I have a scratch build up (only available for a week from this posting): http://koji.fedoraproject.org/koji/taskinfo?taskID=317091
Thanks for confirming, I'll do a formal build and push out an update before closing this.
Fix is in libvirt-0.4.0-2.fc8 shortly hitting updates-testing
This was fixed with that update long ago... closing.