testing/os-autoinst-distri-rocky-migrated

Fork 0

Add minimal and desktop false to all incorrect tests in universal #125

Merged

lumarel merged 1 commit from fix/use-correct-package-set into 9.0_release_fixes

2022-07-23 04:29:06 +00:00

lumarel commented

2022-07-10 21:19:08 +00:00

(Migrated from github.com)

Description

This adds "PACKAGE_SET": "minimal" and "DESKTOP": false where needed to get the tests passing again.

Fixes #81

How Has This Been Tested?

openqa-cli api -X POST isos ISO=Rocky-9.0-x86_64-dvd.iso ARCH=x86_64 DISTRI=rocky FLAVOR=universal VERSION=9.0 BUILD=-universal-$(git branch --show-current)-$(date +%Y%m%d.%H%M%S).0 DNF_CONTENTDIR=true

There are a lot of tests passing now! And a few new issues but mainly a lot of stuff passing! (nothing fails that passed before)

Checklist:

My code follows the style guidelines of this project
I have performed a self-review of my own code
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings
Any dependent changes have been merged and published in downstream modules

# Description This adds `"PACKAGE_SET": "minimal"` and `"DESKTOP": false` where needed to get the tests passing again. Fixes #81 # How Has This Been Tested? ``` openqa-cli api -X POST isos ISO=Rocky-9.0-x86_64-dvd.iso ARCH=x86_64 DISTRI=rocky FLAVOR=universal VERSION=9.0 BUILD=-universal-$(git branch --show-current)-$(date +%Y%m%d.%H%M%S).0 DNF_CONTENTDIR=true ``` There are a lot of tests passing now! And a few new issues but mainly a lot of stuff passing! (nothing fails that passed before) ![image](https://user-images.githubusercontent.com/42647570/178162448-d15dc3bc-097b-49e9-b2ca-246b9e0c070a.png) # Checklist: - [x] My code follows the style guidelines of this project - [x] I have performed a self-review of my own code - [ ] I have commented my code, particularly in hard-to-understand areas - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [x] Any dependent changes have been merged and published in downstream modules

tcooper commented

2022-07-10 23:23:54 +00:00

(Migrated from github.com)

@lumarel You don't indicated if you are using DNF_CONTENTDIR and/or supplying alternate LOCATION variable for install_repository_http_graphical / install_repository_http_variation in How has this been tested? Did you run this FLAVOR=universal exactly as you've shown above?

@lumarel You don't indicated if you are using `DNF_CONTENTDIR` and/or supplying alternate `LOCATION` variable for `install_repository_http_graphical / install_repository_http_variation` in **How has this been tested?** Did you run this `FLAVOR=universal` exactly as you've shown above?

tcooper commented

2022-07-11 05:48:38 +00:00

(Migrated from github.com)

Another observation... While looking at a recent Rawhide build in the Fedora openQA I noticed that many (most?) tests are using Fedora Custom Operating System as the installation environment. The test settings indicated PACKAGE_SET=default and SUBVARIANT=Everything and in most cases do not supply DESKTOP=<anything>.

Such tests end up executing _console_wait_login. I think the different meaning of PACKAGE_SET=default on the Rocky media and lack of SUBVARIANTs is causing much of our login issues.

I am 100% on board with solving that by moving away from using PACKAGE_SET=default and assuming/inferring behavior in favor of explicitly specifying PACKAGE_SET=minimal and DESKTOP=false.

With the possible / (likely) removal of the minimal ISO from Rocky 9 release products this makes even more sense as FLAVOR=minimal-iso was the only other place we were explicitly setting PACKAGE_SET=minimal.

Currently running through a few of these tests above to get confirmation on to get/give the 👍🏻 .

Another observation... While looking at a recent Rawhide build in the Fedora openQA I noticed that many (most?) tests are using `Fedora Custom Operating System` as the installation environment. The test settings indicated `PACKAGE_SET=default` and `SUBVARIANT=Everything` and in most cases do not supply `DESKTOP=<anything>`. Such tests end up executing `_console_wait_login`. I think the different meaning of `PACKAGE_SET=default` on the Rocky media and lack of `SUBVARIANT`s is causing much of our login issues. I am 100% on board with solving that by moving away from using `PACKAGE_SET=default` and assuming/inferring behavior in favor of explicitly specifying `PACKAGE_SET=minimal` and `DESKTOP=false`. With the possible / (likely) removal of the minimal ISO from Rocky 9 release products this makes even more sense as `FLAVOR=minimal-iso` was the only other place we were explicitly setting `PACKAGE_SET=minimal`. Currently running through a few of these tests above to get confirmation on to get/give the 👍🏻 .

👍 2

lumarel commented

2022-07-11 06:32:37 +00:00

(Migrated from github.com)

@lumarel You don't indicated if you are using DNF_CONTENTDIR and/or supplying alternate LOCATION variable for install_repository_http_graphical / install_repository_http_variation in How has this been tested? Did you run this FLAVOR=universal exactly as you've shown above?

Yes you are right there is a difference between what I ran and what I have written above, altered the command now, as the DNF_CONTENTDIR parameter was present but no location 👍🏻
I'm sorry for that, it took me like 7 hours to complete universal, already forgot what I did there ^^

> @lumarel You don't indicated if you are using `DNF_CONTENTDIR` and/or supplying alternate `LOCATION` variable for `install_repository_http_graphical / install_repository_http_variation` in **How has this been tested?** Did you run this `FLAVOR=universal` exactly as you've shown above? Yes you are right there is a difference between what I ran and what I have written above, altered the command now, as the `DNF_CONTENTDIR` parameter was present but no location 👍🏻 I'm sorry for that, it took me like 7 hours to complete universal, already forgot what I did there ^^

lumarel commented

2022-07-11 06:34:12 +00:00

(Migrated from github.com)

And yes this is only a intermediate change to get the stuff working again, it will take me a few weeks even months until we (I will take the task as proposed so I) have changed that to a better suiting structure 👍🏻

tcooper (Migrated from github.com) approved these changes 2022-07-11 15:21:23 +00:00

tcooper (Migrated from github.com) left a comment

Visual inspect seems fine. Also tested install_repository_http_* directly with both successful.

Visual inspect seems fine. Also tested `install_repository_http_*` directly with both successful.

akatch (Migrated from github.com) requested changes 2022-07-17 20:40:41 +00:00

akatch (Migrated from github.com) left a comment

I believe install_sata needs a similar treatment - it fails at console_wait_login but with the GUI login screen shown. Otherwise this looks great!

I believe `install_sata` needs a similar treatment - it fails at `console_wait_login` but with the GUI login screen shown. Otherwise this looks great!

lumarel commented

2022-07-23 00:24:45 +00:00

(Migrated from github.com)

@akatch Added the missing bits for install_sata now 🙂

@akatch Added the missing bits for `install_sata` now 🙂

👍 1

lumarel commented

2022-07-23 01:17:58 +00:00

(Migrated from github.com)