action #36117
closed[functional][u][sporadic] test fails in xterm to show "xterm" (needle tag desktop-runner-plasma-suggestions) in krunner - system slower just after login?
0%
Description
Observation¶
openQA test in scenario opensuse-Tumbleweed-DVD-x86_64-update_13.2@64bit fails in
xterm
to show the "xterm" suggestion in krunner (in time). The test module "xterm" is scheduled just after "user_gui_login" so we try to open krunner shortly after logging in where the system naturally is still more busy than in a later step when we the session is idle. This might be a problem even more in the case of upgrades, maybe because some older data is automatically migrated in user services, e.g. in plasma.
Reproducible¶
Fails sporadically, e.g. 8 jobs ago in the same scenario: https://openqa.opensuse.org/tests/668678#step/xterm/4
Expected result¶
Upgrade tests pass the krunner step in plasma x11 tests in a stable way
Suggestions¶
- Find a smaller reproducer, e.g. load from snapshot before GUI session login, load xterm after that
- Just wait longer in general in desktop-runner, especially when it is krunner
- Alternative: Make sure to wait longer after the login, e.g. in user_gui_login, open krunner (or generic desktop-runner), wait for it show the suggestion and exit without executing the command (without 'ret' press) -> Con: Potentially complicated logic to repeat what other tests do anyway with
x11_start_program
- Alternative: Just wait longer in xterm because it most likely is the first test module after login -> Con: Makes modules more dependant on order
Further details¶
Always latest result in this scenario: latest
Updated by okurz almost 7 years ago
Updated by okurz almost 7 years ago
- Related to coordination #35302: [qe-core][opensuse][functional][epic][sporadic] Various unstable tests on o3 added
Updated by okurz almost 7 years ago
still failed, https://openqa.opensuse.org/tests/677126
What I am thinking is a new explicit test module "x11/desktop_runner" that calls just the desktop runner itself and closes again which might be enough to let the system come back to idle. Then in the next test modules we should be able to rely on the desktop runner without quirks.
Updated by okurz almost 7 years ago
Updated by mgriessmeier almost 7 years ago
- Blocks action #30805: [functional][opensuse][leap][medium][u] first test after reboot fails in krunner, potential system overload (was: test fails in inkscape - typing too fast?) added
Updated by okurz almost 7 years ago
- Status changed from In Progress to Resolved
Updated by okurz almost 7 years ago
- Related to action #37003: [opensuse][functional][u][sporadic] test fails in network_configuration - xterm does not start added
Updated by okurz almost 7 years ago
- Target version changed from Milestone 17 to Milestone 17