What were you trying to do that didn't work?

Migrate vm from 10.0 to 10.1 on GH with postcopy

Please provide the package NVR for which the bug is seen:

Source:
qemu-kvm-9.1.0-15.el10_0.4.aarch64
edk2-aarch64-20241117-2.el10_0.1.noarch
6.12.0-55.33.1.el10_0.aarch64+64k
Dest:
qemu-kvm-10.0.0-13.el10_1.aarch64
edk2-aarch64-20250523-2.el10.noarch
6.12.0-124.el10.aarch64+64k

How reproducible is this bug?:

Steps to reproduce

1. Setup migration env
2. Run migration command

   virsh migrate --live --p2p --persistent --undefinesource --bandwidth 1000 --postcopy --xml /var/tmp/xml_utils_temp_ibehzvxi.xml --persistent-xml /var/tmp/xml_utils_temp_fi27x1wd.xml --domain avocado-vt-vm1 --desturi qemu+tcp://<dest_ip>/system

1. Expected results

   Migration is successful

   Actual results

   qemu-kvm crashed on dest host.
   Qemu log on source:

   2025-09-12 01:59:24.913+0000: initiating migration
   2025-09-12T01:59:28.539780Z qemu-kvm: failed to save SaveStateEntry with id(name): 3(ram): -5
   2025-09-12T01:59:28.539888Z qemu-kvm: Unable to shutdown socket: Transport endpoint is not connected
   2025-09-12T01:59:28.539923Z qemu-kvm: Unable to shutdown socket: Bad file descriptor
   2025-09-12T01:59:28.539947Z qemu-kvm: Detected IO failure for postcopy. Migration paused.
   2025-09-12T02:00:14.206350Z qemu-kvm: terminating on signal 15 from pid 258716 (/usr/sbin/virtqemud)
   2025-09-12 02:00:15.006+0000: shutting down, reason=destroyed

   Qemu log on target:

   2025-09-12 01:59:24.387+0000: 220528: info : hostname: nvidia-grace-hopper-03.khw.eng.rdu2.dc.redhat.com
   2025-09-12 01:59:24.387+0000: 220528: info : virObjectUnref:378 : OBJECT_UNREF: obj=0xffff28007900
   char device redirected to /dev/pts/7 (label charserial0)
   2025-09-12T01:59:28.539463Z qemu-kvm: error while loading state for instance 0x0 of device 'cpu'
   2025-09-12T01:59:28.539610Z qemu-kvm: error while loading state section id 3(ram)
   qemu-kvm: ../util/oslib-posix.c:247: void qemu_socket_set_nonblock(int): Assertion `f == 0' failed.
   2025-09-12 01:59:28.928+0000: shutting down, reason=failed

log:
Running '/bin/virsh -c 'qemu:///system' migrate-postcopy avocado-vt-vm1'
Command '/bin/virsh -c 'qemu:///system' migrate-postcopy avocado-vt-vm1' finished with 0 after 0.022879207s
[stderr] error: internal error: QEMU unexpectedly closed the monitor (vm='avocado-vt-vm1'): 2025-09-12T01:59:28.539463Z qemu-kvm: error while loading state for instance 0x0 of device 'cpu'
[stderr] 2025-09-12T01:59:28.539610Z qemu-kvm: error while loading state section id 3(ram)
[stderr] qemu-kvm: ../util/oslib-posix.c:247: void qemu_socket_set_nonblock(int): Assertion `f == 0' failed.
[stdlog] 2025-09-11 21:59:29,543 avocado.utils.process process L0714 INFO | Command '/bin/virsh -c 'qemu:///system' migrate --live --p2p --persistent --undefinesource --bandwidth 1000 --postcopy --xml /var/tmp/xml_utils_temp_ibehzvxi.xml --persistent-xml /var/tmp/xml_utils_temp_fi27x1wd.xml --domain avocado-vt-vm1 --desturi qemu+tcp://10.6.12.65/system' finished with 1 after 5.262243140s

See attachment for core file.

Auto case: (1/1) type_specific.io-github-autotest-libvirt.migrate.migration_with_numa_topology.with_back_migration.postcopy.base_options.multi_cluster_on_numa: FAIL: error: internal error: QEMU unexpectedly closed the monitor (vm='avocado-vt-vm1'): 2025-09-12T01:59:28.539463Z qemu-kvm: error while loading state for instance 0x0 of device 'cpu'\n2025-09-12T01:59:28.539610Z qemu-kvm: error while loading state section id ... (250.48 s)