QA » openQA Project » openQA Tests

Did you receive feedback because of the SLE bug and the reason that the test is failing (now in Leap)?

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1684566

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1693671

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1715763

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1737114

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released"
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The label in the openQA scenario is removed

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/1757541

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The label in the openQA scenario is removed

@binary_sequence: I know that you do not work for SUSE any more. This test bug exists continuously in our Leap release.
Can you give me some hints, how I can receive more SUSE internal information, what is going wrong here? Then I can fix it for my own.

Hi Sarah,

On Tumbleweed x86_64¶

An installation selecting a desktop on system role, should set systemd graphical.target.
https://openqa.opensuse.org/tests/2059915#step/installation_overview/4
An installation selecting a headless system role, should set systemd multi-user.target.
https://openqa.opensuse.org/tests/2059514#step/installation_overview/4

On Tumbleweed s390x, it is tricky¶

An installation selecting a headless system role, should set systemd multi-user.target
Remote SSH installation: https://openqa.opensuse.org/tests/2054786#step/installation_overview/5

But when installing via VNC
https://openqa.opensuse.org/tests/2054785#step/bootloader_s390/27
Even if a headless system role is selected, the default systemd target is graphical.target.
That is because it is expected that the user will also access the installed system via VNC, so VNC server needs to be started on system boot. In (open)SUSE, VNC server is started on boot by graphical target.
So, on s390x the graphical.target doesn't start Xserver or Wayland, it just starts VNC server.

The check of systemd target is skipped for tumbleweed
https://openqa.opensuse.org/tests/2054785#step/installation_overview/3
Because check just relied on DESKTOP setting to decide if target should be multi-user or graphical.
In that test, DESKTOP=textmode, so the check will expect multi-user.target, but installation is done via remote VNC, so graphical.target is actually set.
The check is skipped with setting BSC1167736=yes.

On Leap s390x¶

The installation is remove VNC
https://openqa.opensuse.org/tests/2054206#step/bootloader_s390/27
So, the systemd target is graphical.target as expected
https://openqa.opensuse.org/tests/2054206#step/installation_overview/5
But, the check is not taking this special case into account.

A quick 'fix'¶

for Leap would be to also add the setting BSC1167736=yes so the check is skipped.

To not interfere with other arquitectures, it should be put on the job group template for test suite textmode-server
https://openqa.opensuse.org/admin/job_templates/50

Similar to how it was done for the job group template for Tumbleweed (search textmode-server)
https://openqa.opensuse.org/admin/job_templates/34

A proper solution¶

would be to update the check to also take into account a remote VNC installtion on s390x and don't skip the check.

Since github is malfunctioning, I cannot show you the part where it is done. But, at least I could get:

• the function where the check is done: https://github.com/os-autoinst/os-autoinst-distri-opensuse/blob/cae6400303ef59c58de143fbaf3050bf5e22457e/tests/installation/installation_overview.pm#L61
• the PR when the skip was done: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/10406
• the ticket for which the skip was done: https://progress.opensuse.org/issues/65301
• the bug is private: https://bugzilla.opensuse.org/show_bug.cgi?id=1167736

I am still interested in openSUSE and s390x. So, do not hesitate to ask if you need my help.

Have fun!

Oh, the openQA modules links don't link to github anymore so: https://openqa.opensuse.org/tests/2054206/modules/installation_overview/steps/1/src

sub check_default_target {
     # Check the systemd target where scenario make it possible
     return if (is_microos || is_sle_micro || is_upgrade || is_hyperv ||
         get_var('REMOTE_CONTROLLER') || (get_var('BACKEND', '') =~ /spvm|pvm_hmc|ipmi/));
     # exclude non-desktop environment and scenarios with edition of package selection (bsc#1167736)
     return if (!get_var('DESKTOP') || get_var('PATTERNS'));
     return if (get_var 'BSC1167736');

     # Set expectations
     my $expected_target = check_var('DESKTOP', 'textmode') ? "multi-user" : "graphical";

     select_console 'install-shell';

     my $target_search = 'default target has been set';
     # default.target is not yet linked, so we parse logs and assert expectations
     if (my $log_line = script_output("grep '$target_search' /var/log/YaST2/y2log | tail -1",
             proceed_on_failure => 1)) {
         $log_line =~ /$target_search: (?<current_target>.*)/;
         assert_equals($expected_target, $+{current_target}, "Mismatch in default.target");
     }

     select_console 'installation';
 }

• The check is skipped with: return if (get_var 'BSC1167736');
• The check is: my $expected_target = check_var('DESKTOP', 'textmode') ? "multi-user" : "graphical";

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/2064455

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The bugref in the openQA scenario is removed or replaced, e.g. label:wontfix:boo1234

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/2105660

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The bugref in the openQA scenario is removed or replaced, e.g. label:wontfix:boo1234

This is an autogenerated message for openQA integration by the openqa_review script:

This bug is still referenced in a failing openQA test: textmode-server
https://openqa.opensuse.org/tests/2128930

To prevent further reminder comments one of the following options should be followed:

1. The test scenario is fixed by applying the bug fix to the tested product or the test is adjusted
2. The openQA job group is moved to "Released" or "EOL" (End-of-Life)
3. The bugref in the openQA scenario is removed or replaced, e.g. label:wontfix:boo1234

I have added "BSC1167736: yes" to the Leap project config now. That did not help in the Leap config.
I have removed the line with the bug reference number, that it won't be skipped any more and expanded the expectations with a star as a regex:
my $expected_target = check_var('DESKTOP', 'textmode*') ? "multi-user" : "graphical";

Pull Request: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/14081