Can you please retest this with current libvirt build? Thanks.

Hi Jaroslav,

Liping(lcheng) gave me the auto case url: https://libvirt-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/view/libvirt/view/RHEL-9.2%20x86_64/job/libvirt-RHEL-9.2-runtest-x86_64-function-migration_priority_3/32/testReport/rhel.virsh/migrate_options_shared/negative_test_vtpm_emulator_without_postcopy/

The latest error msg changed to:
"error: operation failed: job 'migration out' unexpectedly failed"

And per my manual test, besides this error output, target qemu log got same detailed msgs as previous.
#  virsh migrate avocado-vt-vm1 qemu+ssh://hostB/system --live  --verbose
Migration: [ 97 %]error: operation failed: job 'migration out' unexpectedly failed

[hostB]# tail -4 /var/log/libvirt/qemu/avocado-vt-vm1.log 
2023-03-30T03:28:22.430806Z qemu-kvm: tpm-emulator: Setting the stateblob (type 1) failed with a TPM error 0x1f 
2023-03-30T03:28:22.430841Z qemu-kvm: error while loading state for instance 0x0 of device 'tpm-emulator'
2023-03-30T03:28:22.433638Z qemu-kvm: load of migration failed: Input/output error
2023-03-30 03:28:22.637+0000: shutting down, reason=crashed

So the current error output is still somewhat vague. Is it possible to print specific msgs since they're already there in qemu log?

Thanks!

This feels like we're overwriting error message somewhere in cleanup path after failed migration. But unfortunately, I'm unable to reproduce the error during migration. Is there anything special I need to do?

(In reply to Michal Privoznik from comment #3)
> But unfortunately, I'm unable to reproduce the error
> during migration. Is there anything special I need to do?

Hi Michal,
Just try vtpm secrets with different values please:
host A:
# MYSECRET=`printf %s "open sesame" | base64`
# virsh secret-set-value 6dd3e4a5-1d76-44ce-961f-f119f5aad935 $MYSECRET
Secret value set

hostB:
# MYSECRET=`printf %s "my abc" | base64`
# virsh secret-set-value 6dd3e4a5-1d76-44ce-961f-f119f5aad935 $MYSECRET
Secret value set

This is a general migration issue not connected to vtpm at all. It's a race
between Finish migration phase and a cleanup code that triggers when QEMU
process dies. If libvirt on the destination host cleans everything up before
the source (for p2p, or a client for non-p2p migration), the destination just
reports the domain does not exist anymore.

Fixing it will be nice, although we'd need to come up with a clever solution
to avoid keeping a dead domain present forever in case Finish is never called.

Issue migration from Bugzilla to Jira is in process at this time. This will be the last message in Jira copied from the Bugzilla bug.

This BZ has been automatically migrated to the issues.redhat.com Red Hat Issue Tracker. All future work related to this report will be managed there.

Due to differences in account names between systems, some fields were not replicated.  Be sure to add yourself to Jira issue's "Watchers" field to continue receiving updates and add others to the "Need Info From" field to continue requesting information.

To find the migrated issue, look in the "Links" section for a direct link to the new issue location. The issue key will have an icon of 2 footprints next to it, and begin with "RHEL-" followed by an integer.  You can also find this issue by visiting https://issues.redhat.com/issues/?jql= and searching the "Bugzilla Bug" field for this BZ's number, e.g. a search like:

"Bugzilla Bug" = 1234567

In the event you have trouble locating or viewing this issue, you can file an issue by sending mail to rh-issues. You can also visit https://access.redhat.com/articles/7032570 for general account information.