Update 1 cloud-init case per rebase build new change

[zhaohuijuan]

Update the case per RHEL-17889 - [rhel-9] Align with upstream to standardise module names in cloud.cfg to only use underscore

[zhaohuijuan]

Update another check for "cloud-init status" per the new change in cloud-init-23.4.
Do not check the return code(0), only check the output, per the change as below:

Cloud-init error codes
----------------------
0 - success
1 - unrecoverable error
2 - recoverable error (new)

This new exit code indicates recoverable errors. If cloud-init exits
with exit code (2), cloud-init was able to complete gracefully, however
something went wrong and the user should investigate.

[liangxiao1]

How about collecting "cloud-init status --format json" output when ret code does not equal 0? I am also thinking , can we fail the case when ret code is not 0? Otherwise, we might miss bugs when it returns 1 or 2(also require fix?).

[zhaohuijuan]

Good idea and it makes sense.
This funtion _check_cloudinit_done_and_service_isactive() is called by 10 cases, the return code 2 is always no relation with the 10 cases, and all the 10 cases would fail when return code is not 0, this will add the debug effort for failed cases.
How about adding a new case to check the "cloud-init status" return code? We can fail the case when ret code is not 0, and we can collect "cloud-init status --format json" output when ret code does not equal 0 in this new case.

[liangxiao1]

When return code is non-zero, it is not critical some times, I agree to add a new case to catch non-zero ret. I think it is worth to do because "cloud-init status --format json" indicates there are some exceptions found.

[zhaohuijuan]

Thanks Frank for the suggestion, merging now and will add a new case to catch non-zero return ASAP.

[xiachen-rh]

LGTM