On my machine, the F17 jeos image leaves an apparently harmless kernel
backtrace in dmesg:

https://bugzilla.redhat.com/show_bug.cgi?id=921724

This causes the qemu migration tests to claim failure, even though the guest
is responsive. Here's the stack:

15:03:04 ERROR| Traceback (most recent call last):
15:03:04 ERROR|   File
"/home/crobinso/src/autotest-virt-tests/virttest/standalone_test.py", line
195, in run_once
15:03:04 ERROR|     run_func(self, params, env)
15:03:04 ERROR|   File
"/home/crobinso/src/autotest-virt-tests/qemu/tests/migration.py", line 78, in
run_migration
15:03:04 ERROR|     check)
15:03:04 ERROR|   File
"/usr/lib/python2.7/site-packages/autotest/client/shared/error.py", line 138,
in new_fn
15:03:04 ERROR|     return fn(*args, **kwargs)
15:03:04 ERROR|   File
"/home/crobinso/src/autotest-virt-tests/virttest/qemu_vm.py", line 2732, in
migrate
15:03:04 ERROR|     self.verify_alive()
15:03:04 ERROR|   File
"/home/crobinso/src/autotest-virt-tests/virttest/qemu_vm.py", line 114, in
verify_alive
15:03:04 ERROR|     self.verify_kernel_crash()
15:03:04 ERROR|   File
"/home/crobinso/src/autotest-virt-tests/virttest/virt_vm.py", line 707, in
verify_kernel_crash
15:03:04 ERROR|     raise VMDeadKernelCrashError(match.group(0))


Certainly it's useful to catch these errors even if the guest is responsive,
but maybe it shouldn't be unconditionally fatal? Or only do the check after it
appears the guest is wedged?

Anyone have any thoughts?

Thanks,
Cole

_______________________________________________
Virt-test-devel mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/virt-test-devel

Reply via email to