action #69046
closed
coordination #42854: [functional][y][epic] Introduce relevant installation validation test modules
[y] Add validation to lvm+cancel_existing_cryptlvm
Added by riafarov almost 4 years ago.
Updated over 3 years ago.
Target version:
SUSE QA - SLE 15 SP3
Description
See motivation in the parent ticket.
This scenario is special about cancelling enabling encrypted partitions. It doesn't make sense to validate this after installation is conducted, as disk will be wiped out.
Nevertheless, we can check that existing encrypted partitions are NOT activated. This should be done in #69040
We already have tests/console/validate_lvm.pm
and tests/console/validate_encrypted_partition_not_activated.pm
which we can reuse for the validations.
Acceptance criteria¶
- Ignorance of the encrypted partitions is validated
- Partitioning in the installed system is validated
- Test plan is updated accordingly
- Due date changed from 2020-08-25 to 2020-08-11
- Target version set to SLE 15 SP3
- Due date changed from 2020-08-11 to 2020-08-25
- Due date changed from 2020-08-25 to 2020-09-08
- Description updated (diff)
- Status changed from New to Workable
- Estimated time set to 5.00 h
- Due date changed from 2020-09-08 to 2020-09-22
- Description updated (diff)
- Status changed from Workable to Feedback
- Due date changed from 2020-09-22 to 2020-10-06
- Status changed from Feedback to In Progress
The PR did not work for PVM. So it reminded me of that other PR: https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/10786. So I took the solution from that one and it works. But also in that other PR, we see that I had used a different device name to use for PVM : cr-auto-1 instead od cr-auto-2. I cannot remember why I had to do that, but I remember it was essential and made sense.
Now we have a problem as currently, the module validate_encrypted_partition_not_activated does "cryptsetup status /dev/mapper/cr-auto-2" and expects to have as a result "inactive"
but if I write "cryptsetup status my_grandma", it will also succeed as the output will be "my_grandma is inactive".
So I'm trying to figure out again what is the proper device to use, and at least comment out how tho find it, as I completely forgot how I found it out back in august.
- Status changed from In Progress to Feedback
I managed to get this information the "hard way" (see description of PR) so we need to use a different device name for PVM. I will propose a follow-up to improve this case of potential false positive.
- Status changed from Feedback to Resolved
Also available in: Atom
PDF