Adjust yast control center for sle15 #3916
Conversation
| @@ -34,6 +34,7 @@ sub search { | |||
| } | |||
| else { | |||
| send_key 'alt-s'; | |||
| send_key 'backspace'; | |||
There was a problem hiding this comment.
This part doesn't work as expected, as when this method called with empty string, we expect that field is emptied then. But with old code we just navigate to the search field.
| @@ -136,13 +154,13 @@ sub start_authentication_server { | |||
| assert_and_click 'yast2_control-center_authentication-server'; | |||
| assert_screen [qw(yast2_control-center-authentication-server_install yast2_control-center-authentication-server_configuration)], 90; | |||
| if (match_has_tag('yast2_control-center-authentication-server_install')) { | |||
| send_key 'alt-i'; | |||
| wait_still_screen { send_key 'alt-i' }; | |||
There was a problem hiding this comment.
why calling wait_still_screen when there is an assert_screen just afterwards?
| assert_screen 'yast2_control-center-authentication-server_configuration'; | ||
| send_key 'alt-c'; | ||
| wait_screen_change { send_key 'alt-c' }; |
There was a problem hiding this comment.
same here, why wait_screen_change when there is an assert_screen just afterwards?
rwx788
force-pushed
the
27074_yast2gui
branch
from
d462f05
to
e76c7ec
Compare
Originally we thought we just need to create new set of needles, but some of modules do not behave as expected or are not pre-installed. This commit includes additional soft-failures and some minor improvements. See [poo#25716](https://progress.opensuse.org/issues/25716).
rwx788
force-pushed
the
27074_yast2gui
branch
from
e76c7ec
to
1a5d689
Compare
|
NOTE: please merge needles for bootloader as well. Needles for control center already merged. |
|
needles merged. @Zaoliang can you take a look and review the code? |
| assert_screen 'yast2_control-center-partitioner_expert', timeout => 60; | ||
| send_key 'alt-f'; | ||
| if (is_storage_ng) { | ||
| assert_screen 'expert-partitioner-modify-confirmation'; | ||
| wait_screen_change { send_key 'alt-o' }; #Continue |
There was a problem hiding this comment.
that wait_screen_change again shouldn't be needed. We have a clear assert_screen afterwards. That's enough, isn't it?
| @@ -273,6 +291,12 @@ sub run { | |||
| if (is_sle && sle_version_at_least '15') { | |||
| #see bsc#1062331, sound is not added to the yast2 pattern | |||
| ensure_installed('in yast2-sound'); | |||
| #kdump is disabled by default, so ensure that it's installed | |||
| ensure_installed('in yast2-kdump'); | |||
There was a problem hiding this comment.
how is "yast2-kdump" related to the comment that "kdump is disabled by default"? Btw, it shouldn't be disabled by default because as far as I remember at least on a default SLES installation it will enable kdump during installation, isn't it?
There was a problem hiding this comment.
kdump is disabled in installation by default, hence nothing is installed and this is expected. In the qcow image used here it's disabled and we lack defined flow how to be sure if it's there or not in given image.
There was a problem hiding this comment.
right, it's disabled by default. I also checked SLE 12-SP3, nevermind
| @@ -273,6 +291,12 @@ sub run { | |||
| if (is_sle && sle_version_at_least '15') { | |||
| #see bsc#1062331, sound is not added to the yast2 pattern | |||
| ensure_installed('in yast2-sound'); | |||
| #kdump is disabled by default, so ensure that it's installed | |||
| ensure_installed('in yast2-kdump'); | |||
| record_soft_failure 'bsc#1059569'; | |||
There was a problem hiding this comment.
Would it work to check for the error condition first? E.g. with multi-tag assert_screen and match_has_tag?
There was a problem hiding this comment.
Which error condition? We can verify that package is not installed, then soft-fail and install, or see that module is not there -> install -> restart control center, but that looks like an overhead for this case. We can also just fail on the bug and see which decision is made.
| } | ||
| else { | ||
| start_wake_on_lan; | ||
| } |
There was a problem hiding this comment.
IMHO it looks cleaner if that is handled within the start_wake_on_lan method. Also this should allow to check for the error condition first as well
There was a problem hiding this comment.
Right, here we can do proper soft-fail without workaround: https://gitlab.suse.de/openqa/os-autoinst-needles-sles/merge_requests/579 Please, merge new needle.
| send_key 'alt-y'; | ||
| assert_screen [qw(yast2_control-center-partitioner_warning yast2_control-center-partitioner_expert)], 180; | ||
| # Define if storage-ng | ||
| set_var('STORAGE_NG', 1) if match_has_tag 'storage-ng'; |
There was a problem hiding this comment.
not sure ... I would prefer to keep decision maker as a human for this variable. It's too important to leave it like this
There was a problem hiding this comment.
We have the same approach in the installer, so for this test suite variable is not set. Of course we can also set it on all mediums, but this line of code doesn't hurt anyway.
There was a problem hiding this comment.
@asmorodskyi there is a long history behind that variable. Simply stated: We won't go back anytime soon :)
rwx788
force-pushed
the
27074_yast2gui
branch
from
1a5d689
to
add2c07
Compare
|
With latest enhancement to wake on lan we need one needle for sle: |
In the test we check value in dropdown, but then it overlaps other controls. Pressing esc key closes it.
rwx788
force-pushed
the
27074_yast2gui
branch
from
add2c07
to
bb3c3b0
Compare
|
all needles PRs/MRs merged |
Originally we thought we just need to create new set of needles, but
some of modules do not behave as expected or are not pre-installed. This
commit includes additional soft-failures and some minor improvements.