Dear Maintainer,

   * What led up to the situation?
     Upgrading Debian/testing
   * What exactly did you do (or not do) that was effective (or
     ineffective)?
     # apt -u -V dist-upgrade
   * What was the outcome of this action?
     None of my auto-start VM's boot with the system comes up
   * What outcome did you expect instead?
     All of my auto-start VM's get executed while the system comes up

misabnormaly described above.

Here the problem that shows up with journalctl:

Jun 11 06:53:23 elitedesk.lan.dac systemd[1]: Starting libvirtd.service -
libvirt legacy monolithic daemon...
Jun 11 06:53:25 elitedesk.lan.dac systemd[1]: Started libvirtd.service -
libvirt legacy monolithic daemon.
Jun 11 06:53:28 elitedesk.lan.dac systemd-coredump[4510]: [🡕] Process 2952
(libvirtd) of user 0 dumped core.

                                                          Module
libnss_systemd.so.2 from deb systemd-260.1-1.amd64
                                                          Module libuuid.so.1
from deb util-linux-2.42.1-3.amd64
                                                          Module libudev.so.1
from deb systemd-260.1-1.amd64
                                                          Module
libsystemd.so.0 from deb systemd-260.1-1.amd64
                                                          Module libblkid.so.1
from deb util-linux-2.42.1-3.amd64
                                                          Module libzstd.so.1
from deb libzstd-1.5.7+dfsg-3+b2.amd64
                                                          Module libmount.so.1
from deb util-linux-2.42.1-3.amd64
                                                          Module libatomic.so.1
from deb gcc-16-16.1.0-1.amd64
                                                          Module libgcc_s.so.1
from deb gcc-16-16.1.0-1.amd64
                                                          Stack trace of thread
4254:
                                                          #0
0x00007f5b9c2eed15 n/a (libvirt_driver_qemu.so + 0xeed15)
                                                          #1
0x00007f5b9c2f7377 n/a (libvirt_driver_qemu.so + 0xf7377)
                                                          #2
0x00007f5b9c2f77f3 n/a (libvirt_driver_qemu.so + 0xf77f3)
                                                          #3
0x00007f5ba4056b65 n/a (libvirt.so.0 + 0x256b65)
                                                          #4
0x00007f5ba3fd3bde n/a (libvirt.so.0 + 0x1d3bde)
                                                          #5
0x00007f5ba3f0994d virHashForEachSafe (libvirt.so.0 + 0x10994d)
                                                          #6
0x00007f5ba3fd531b virDomainObjListForEach (libvirt.so.0 + 0x1d531b)
                                                          #7
0x00007f5ba40585c6 virDomainDriverAutoStart (libvirt.so.0 + 0x2585c6)
                                                          #8
0x00007f5b9c2ca46e n/a (libvirt_driver_qemu.so + 0xca46e)
                                                          #9
0x00007f5ba4145cd7 virStateInitialize (libvirt.so.0 + 0x345cd7)
                                                          #10
0x0000564fbdd33c99 n/a (/usr/sbin/libvirtd + 0x4bc99)
                                                          #11
0x00007f5ba3f52afa n/a (libvirt.so.0 + 0x152afa)
                                                          #12
0x00007f5ba38b4da9 start_thread (libc.so.6 + 0x95da9)
                                                          #13
0x00007f5ba3933e08 __clone3 (libc.so.6 + 0x114e08)

...

As you can see above, the libvirtd process from root while the system comes up
dumps core.

After the system enters login ready state, I can manually start my auto-start
VM's with virt-manger -- which works good.

Thank you for your attention.

Sincerely,

Adrian Kiess

Dear Maintainer,

I like to make a addition:

I configured my VM guests like this:

root@elitedesk /home/adrian # cat /etc/default/libvirt-guests | grep
ON_SHUTDOWN
#             ON_SHUTDOWN=shutdown, you must also set SHUTDOWN_TIMEOUT to a
#ON_SHUTDOWN=shutdown
ON_SHUTDOWN=suspend

A few days ago, I experienced a power outage and when starting my system
back up, the VM's came up OK with auto-start while now being run as
cold-start and not from suspend.

With resume from suspend previously reported error persists.

Thank you.

Adrian Kiess

Hello dear Maintainer,

I like to make the following addition —

The error I find with journalctl about the first reported segfault is
the following:

root@elitedesk /home/adrian # journalctl -n10000 | grep segfault
Aug 08 06:34:31 elitedesk.lan.dac kernel: daemon-init[4290]: segfault at
88 ip 00007f984b8e9228 sp 00007f9823ffe548 error 4 in
libvirt_driver_qemu.so[e9228,7f984b84c000+13b000] likely on CPU 2 (core
2, socket 0)

Thank you.

Adrian