On 08/24/2012 05:16 AM, Shawn Furrow wrote: > Hi all, > > For some reason when I run "libvirtd -d" as root it seems to not start. I > investigate by running "libvirtd -v" as root and I get a segfault. Here is > the output: > > 2012-08-23 19:01:23.237+0000: 27464: info : libvirt version: 0.10.0 >> 2012-08-23 19:01:23.237+0000: 27464: error : virDriverLoadModule:78 : >> failed to load module >> /usr/local/lib/libvirt/connection-driver/libvirt_driver_qemu.so >> /usr/local/lib/libvirt/connection-driver/libvirt_driver_qemu.so: undefined >> symbol: virSecurityManagerGetProcessLabel >> 2012-08-23 19:01:23.238+0000: 27464: error : virDriverLoadModule:78 : >> failed to load module >> /usr/local/lib/libvirt/connection-driver/libvirt_driver_lxc.so >> /usr/local/lib/libvirt/connection-driver/libvirt_driver_lxc.so: undefined >> symbol: virSecurityManagerGetProcessLabel
Did we manage to get this solved in today's 0.10.1 release, or it is still an issue we need to investigate? > However, when I run "libvirtd -d" as me, it does not segfault and continues > running. However I cannot connect to the libvirt-sock because it gets > created at "$HOME/.cache/libvirt/libvirt-sock" but virsh looks for it at > "/usr/local/var/run/libvirt-sock". This also makes it sound like you may be running into an issue of version mismatch, where depending on _how_ you run libvirtd affects _where_ the .so search path is looking, and where your crashing version seems to be picking up on stale .so files. You may want to double check that you have completely uninstalled any old versions before your next attempt, to avoid issues of stale .so files. -- Eric Blake ebl...@redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@redhat.com https://www.redhat.com/mailman/listinfo/libvir-list
