Am 03.05.2019 um 16:43 schrieb Martin Kletzander:
On Thu, May 02, 2019 at 02:25:28PM +0200, Martin Kletzander wrote:
On Thu, May 02, 2019 at 01:07:32PM +0200, Michal Privoznik wrote:
On 5/2/19 12:08 PM, Lothar Schilling wrote:

I turned logging up to maximum. That's all I get:

May  2 11:02:06 hl308-3 systemd: Starting Virtualization daemon...
May  2 11:02:06 hl308-3 libvirtd: 472: info : libvirt version: 4.5.0, package: 10.el7_6.7 (CentOS BuildSystem <http://bugs.centos.org>, 2019-04-24-14:04:12, x86-01.bsys.centos.org)
May  2 11:02:06 hl308-3 libvirtd: 472: info : hostname: my.server.net
May  2 11:02:06 hl308-3 libvirtd: 472: info : virEventPollAddHandle:140 : EVENT_POLL_ADD_HANDLE: watch=1 fd=6 events=1 cb=0x7eff3bbde300 opaque=(nil) ff=(nil)
May  2 11:02:06 hl308-3 libvirtd: 472: info : virEventPollAddHandle:140 : EVENT_POLL_ADD_HANDLE: watch=2 fd=8 events=1 cb=0x7eff3bd06760 opaque=0x558f9db485a0 ff=(nil)
May  2 11:02:06 hl308-3 systemd: Started Virtualization daemon.
May  2 11:02:06 hl308-3 libvirtd: 472: info : virEventPollAddHandle:140 : EVENT_POLL_ADD_HANDLE: watch=3 fd=11 events=0 cb=0x7eff3bcfb420 opaque=0x558f9db69520 ff=0x7eff3bcfb3d0
May  2 11:02:06 hl308-3 libvirtd: 472: info : virEventPollAddHandle:140 : EVENT_POLL_ADD_HANDLE: watch=4 fd=12 events=0 cb=0x7eff3bcfb420 opaque=0x558f9db69700 ff=0x7eff3bcfb3d0
May  2 11:02:06 hl308-3 libvirtd: process 472: arguments to dbus_message_iter_append_basic() were incorrect, assertion "_dbus_check_is_valid_utf8 (*string_p)" failed in file ../../dbus/dbus-message.c line 2754.
May  2 11:02:06 hl308-3 libvirtd: This is normally a bug in some application using the D-Bus library.
May  2 11:02:06 hl308-3 libvirtd: D-Bus not built with -rdynamic so unable to print a backtrace
May  2 11:02:06 hl308-3 systemd: libvirtd.service: main process exited, code=killed, status=6/ABRT

This is not the maximum loggin. We'll need to see debug logs:

https://wiki.libvirt.org/page/DebugLogs

and perhaps running libvirtd under gdb and getting a stack trace (with
all strings expanded) might help too.


Actually gdb might be the only help here.  The reason is that that dbus call
just reports the error and then calls abort(), so libvirtd will not even get an
error message (which would help us track down what might've been the case).

If you can do something like:


Even better, there are some ways to get the core dump information which I
completely forgot about:

- coredumpctl -- IIRC you are running CentOS 7, so this will not be available
                 for you yet.  If it was, coredumpctl show just shows all we
                 need (by default it picks the last core dump and stacktrace is
                 part of the output).

- abrt-cli -- Similar to the above, it should be available for you, but you
              need to have abrt installed and setup.  Which is not always the
              default.  If you do, or you set it up yourself, however, you
              should be able to get to the core dump as well, but it is not as
              easy as with coredumpctl (i.e. I did not get to it in first 10
              seconds).

- Setup your own -- either set the ulimit for the service and look for the file
                    that gets created or set kernel.core_pattern using sysctl
                    to a command that gets ran with each core dump (this one
                    does not need any ulimit setting, but you need to know what
                    to specify there.

Please let us know if you got anywhere, I'd like to see what the issue is and
how we can fix it.

 $ dnf --enablerepo='*debug*' install libvirt-debuginfo dbus-debuginfo

 $ cat >/var/lib/libvirt/gdbabortscript <<EOF
 start
 break abort
 commands
 t a a bt full
 end
 continue
 EOF

 $ cat >/etc/systemd/system/libvirtd.service.d/override.conf <<EOF
 [Service]
 ExecStart=
 ExecStart=gdb --batch -x /var/lib/libvirt/gdbabortscript /usr/sbin/libvirtd $LIBVIRTD_ARGS
 EOF

 $ systemctl daemon-reload
 $ restorecon -F /var/lib/libvirt/gdbabortscript /etc/systemd/system/libvirtd.service.d/override.conf
 $ systemctl restart libvirtd.service

You should get the full stacktrace of the issue in the output of:

 journalctl -u libvirtd.service

and you can post it here so we can find out what's happening.

HTH,
Martin

Thank you, but I decided to reformat the machine and start from scratch.