action #56441
closed
[sle][functional][y] test fails in partitioning_full_lvm - Volume management did not get expanded
Added by JERiveraMoya over 4 years ago.
Updated over 4 years ago.
Category:
Bugs in existing tests
Target version:
SUSE QA - Milestone 28
Description
Observation¶
openQA test in scenario sle-12-SP5-Server-DVD-s390x-lvm-full-encrypt@s390x-kvm-sle12 fails in
partitioning_full_lvm
Instead of expanding Volume Management menu to be able to check vg-system, it is going directly to the root of the tree, in this case linux and what tries to expand is vda.
Failed 2 out 3 runs.
Test suite description¶
Maintainer: riafarov poo#15926
(crypt-)LVM installations can take longer, especially on non-x86_64 architectures. Test suite creates encrypted lvm partition using expert partitioner and performs installation.
Reproducible¶
Fails since (at least) Build 0303 (current job)
Expected result¶
Last good: 0301 (or more recent)
Further details¶
Always latest result in this scenario: latest
- Description updated (diff)
- Due date set to 2019-10-08
- Target version set to Milestone 28
- Priority changed from Normal to High
This is an autogenerated message for openQA integration by the openqa_review script:
This bug is still referenced in a failing openQA test: lvm-encrypt-separate-boot
https://openqa.suse.de/tests/3368946
- Status changed from New to Workable
- Estimated time set to 5.00 h
- Status changed from Workable to In Progress
It looks like:
- key home is sent
- assert_screen_change sees a screen change where there is not (I suspect a glitch to trigger it)
- it directly matches the next needle "volume_management_feature" before the screen actually had time to change, so no send_key 'down' is sent, because volume_management_feature is still higlighted from previous function "addvg"
- starts the next instruction send_key_until_needlematch('lvm_uncollapse_vgs', 'right') which has effect to go down, expand the disk, then the partiton, and never reaches lvm_uncollapse_vgs.
This happens also on aarch64.
Piece of code:
send_key $cmd{system_view};
assert_screen_change(sub {
send_key 'home';
}, 5);
send_key_until_needlematch('volume_management_feature', 'down');
wait_still_screen(stilltime => 2, timeout => 4);
# Expand collapsed list with VGs
send_key_until_needlematch('lvm_uncollapse_vgs', 'right') if is_sle('<15');
I don't see any glitch on video, So I assume that what happens is: send_key 'home' is sent before send_key $cmd{system_view}; is received. trying some workaround.
- Status changed from In Progress to Feedback
- Status changed from Feedback to Resolved
Also available in: Atom
PDF