QA » openQA Project » openQA Tests

It seems like a needle mismatch issue, but may need more investigation to check if it is a product bug

Reference job: https://duck-norris.qe.suse.de/tests/11889

https://openqa.opensuse.org/tests/3102310#step/amarok/23 is completely white, not sure if a openQA hickup or a real issue. The reference job will show.

Cloned job has issues: https://duck-norris.qe.suse.de/tests/11891#step/updates_packagekit_kde/49

And while waiting for https://openqa.opensuse.org/tests/3103115 it got obsoleted and is now https://openqa.opensuse.org/tests/3103365

This failure is very weird. See https://openqa.opensuse.org/tests/3103365#step/amarok/23 - It does not tell me what the failure is. From the autoinst-log.txt the only thing that somehow point to a failure is

[2023-02-07T16:49:46.891615+01:00] [info] [pid:50479] ::: backend::baseclass::set_tags_to_assert: NO matching needles for linuxrc-install-fail
   at /usr/lib/os-autoinst/basetest.pm line 315.
        basetest::run_post_fail(amarok=HASH(0x56429b8c2ef0), "test amarok failed") called at /usr/lib/os-autoinst/basetest.pm line 379
        basetest::runtest(amarok=HASH(0x56429b8c2ef0)) called at /usr/lib/os-autoinst/autotest.pm line 387
        eval {...} called at /usr/lib/os-autoinst/autotest.pm line 387
        autotest::runalltests() called at /usr/lib/os-autoinst/autotest.pm line 244
        eval {...} called at /usr/lib/os-autoinst/autotest.pm line 244
        autotest::run_all() called at /usr/lib/os-autoinst/autotest.pm line 295
        autotest::__ANON__(Mojo::IOLoop::ReadWriteProcess=HASH(0x56429bd70200)) called at /usr/lib/perl5/vendor_perl/5.26.1/Mojo/IOLoop/ReadWriteProcess.pm line 326
        eval {...} called at /usr/lib/perl5/vendor_perl/5.26.1/Mojo/IOLoop/ReadWriteProcess.pm line 326
        Mojo::IOLoop::ReadWriteProcess::_fork(Mojo::IOLoop::ReadWriteProcess=HASH(0x56429bd70200), CODE(0x56429d801020)) called at /usr/lib/perl5/vendor_perl/5.26.1/Mojo/IOLoop/ReadWriteProcess.pm line 489
        Mojo::IOLoop::ReadWriteProcess::start(Mojo::IOLoop::ReadWriteProcess=HASH(0x56429bd70200)) called at /usr/lib/os-autoinst/autotest.pm line 297
        autotest::start_process() called at /usr/lib/os-autoinst/OpenQA/Isotovideo/Runner.pm line 94
        OpenQA::Isotovideo::Runner::start_autotest(OpenQA::Isotovideo::Runner=HASH(0x56429520de70)) called at /usr/bin/isotovideo line 133

But I'm not sure. Investigating further.

I don't hear any sound in https://openqa.opensuse.org/tests/3103365#step/amarok/22 as opposed to https://openqa.opensuse.org/tests/3101798#step/amarok/22

[2023-02-07T16:48:09.333886+01:00] [debug] [pid:50353] >>> testapi::_handle_found_needle: found amarok-20210309, similarity 0.99 @ 194/172
[2023-02-07T16:48:09.334132+01:00] [debug] [pid:50353] tests/x11/amarok.pm:33 called testapi::assert_recorded_sound
[2023-02-07T16:48:09.334163+01:00] [debug] [pid:50353] <<< basetest::stop_audiocapture()
snd2png: 1 channels, samplerate 44100 Hz, 269875 frames (6.12 seconds)
snd2png: 881 frequency bins
spectogram samples: 610
max amplitude: 0.000000
silences: 610 0
[2023-02-07T16:48:10.039903+01:00] [debug] [pid:50353] >>> basetest::verify_sound_image: failed to find DTMF-159D
[2023-02-07T16:48:10.046678+01:00] [debug] [pid:50353] tests/x11/amarok.pm:35 called testapi::send_key
[2023-02-07T16:48:10.046844+01:00] [debug] [pid:50353] <<< testapi::send_key(key="ctrl-q", wait_screen_change=0, do_wait=0)
[2023-02-07T16:48:10.384528+01:00] [debug] [pid:50353] lib/x11test.pm:27 called testapi::assert_screen
[2023-02-07T16:48:10.384659+01:00] [debug] [pid:50353] <<< testapi::assert_screen(mustmatch="generic-desktop", timeout=30)

Hmmmm, looks like the assert_recorded_sound has some issues. This is very difficult to see.

In https://openqa.opensuse.org/tests/3101798#step/amarok/22 is looks like the following

[2023-02-07T00:20:28.689080+01:00] [debug] [pid:16327] tests/x11/amarok.pm:33 called testapi::assert_recorded_sound
[2023-02-07T00:20:28.689130+01:00] [debug] [pid:16327] <<< basetest::stop_audiocapture()
snd2png: 1 channels, samplerate 44100 Hz, 306534 frames (6.95 seconds)
snd2png: 881 frequency bins
spectogram samples: 694
max amplitude: 106.484187
silences: 17 182
[2023-02-07T00:20:29.292204+01:00] [debug] [pid:16327] >>> basetest::verify_sound_image: found DTMF-159D-20200325, similarity 1.00 @ 95/400

The error originates from here.

I need to setup a simpler test setup to investigate and fix this issue. First will create a qcow2 image of the updates system, and then I can just boot that image with the amarok test run.

Create HDD: https://duck-norris.qe.suse.de/tests/11893#

ARGH. Why can't I just clone the job? It constantly fails and I have the tests and the needles up-to-date.

Finally, step 1: https://duck-norris.qe.suse.de/tests/11901

@ph03nix,

The issue should be caused by my changes to the qcow2 image.

I will try to restore the audio volume configuration as I mentioned in https://progress.opensuse.org/issues/124077

Could you please hold on for a while on this ticket?

rfan1 wrote:

@ph03nix,

The issue should be caused by my changes to the qcow2 image.

I will try to restore the audio volume configuration as I mentioned in https://progress.opensuse.org/issues/124077

Could you please hold on for a while on this ticket?

Thanks for letting me know, of course!

ph03nix wrote:

rfan1 wrote:

@ph03nix,

The issue should be caused by my changes to the qcow2 image.

I will try to restore the audio volume configuration as I mentioned in https://progress.opensuse.org/issues/124077

Could you please hold on for a while on this ticket?

Thanks for letting me know, of course!

The issue is gone after I restored the image, I will move on to fix others.