Package: autopkgtest
Version: 5.55
Severity: normal
(This is most likely not an autopkgtest bug, but I couldn't determine
the root cause, and this affects autopkgtest so I thought it best to
document it here for those affected, until it can be re-assigned.)
Recently, autopkgtests using unstable images freshly built with
--boot=efi fail to run with:
qemu-system-x86_64: terminating on signal 15 from pid 178300
(/usr/bin/python3)
<VirtSubproc>: failure: timed out waiting for 'login prompt on serial
console'
autopkgtest [21:24:02]: ERROR: testbed failure: unexpected eof from the
testbed
Exception ignored in: <_io.TextIOWrapper name=6 encoding='UTF-8'>
BrokenPipeError: [Errno 32] Broken pipe
Booting the VM manually, systemd hangs and times out waiting for the
following services to start up:
(1 of 2) Job dev-mapper-loop0p1.device/start running (3s / no limit)
(2 of 2) Job dracut-initqueue.service/start running (8s / no limit)
I'm not familiar with early boot initrd environments but the 'loop0p1'
seems suspiciously like something that might have bled in from the host
environment, because vmdb2 uses that during image creation.
Importantly, I noticed that a trixie image build was not affected, so
this shouldn't be related to autopkgtest itself.
Steps to reproduce:
$ sudo autopkgtest-build-qemu --boot=efi unstable /tmp/unstable.img
...
$ autopkgtest -B -U <pkg> -- qemu /tmp/unstable.img
Best,
Christian
