Re: Do you plan d-i alpha tests? (Re: Updates to the trixie freeze policy)

Roland Clobus <[email protected]> Sat, 30 Nov 2024 21:02:45 +0100
Newsgroups gmane.linux.debian.devel.release,gmane.linux.debian.devel.boot,gmane.linux.debian.devel.gtk-gnome
Message-ID <[email protected]>
Hello Andreas, lists,

On 30/11/2024 19:28, Andreas Tille wrote:
...
> Additionally, I wonder if an automated test suite for the installer
> might be helpful.
...

There is already some fully automated testing in place.

The d-i installer is used at least in 2 different modes: in the netinst 
image and in the live images. Both modes are used in automated openQA 
tests [1], which performs a few tests, emulating what a user would do 
(i.e. no preseeding):
* Boot the medium and continue until the first screen of the installer 
for each entry in the GRUB/isolinux menu
* Select each language in d-i and continue until the first screen
* Install each available desktop from tasksel (netinst) + boot into that 
image
* Install each desktop image (live) + boot into that image

The installation steps are for a typical desktop environment, assuming 
an empty hard disk.
The existing test can select between an unencrypted and encrypted scenario.

The netinst tests are generated for 'testing' and run 3x per day. All 
live images for 'sid' are tested daily, for 'testing' 2x/week and 
'stable' and 'oldstable' monthly.

The d-i is a very complex product, there are many possible ways (code 
paths) to traverse the software until the end. Surely we can't reach 
100% code coverage easily.

Andreas, what do you have in mind regarding the tests? What scenarios 
would be needed/required?

With kind regards,
Roland Clobus

[1] https://openqa.debian.net/
OpenPGP_signature.asc (application/pgp-signature, 833 B)
-----BEGIN PGP SIGNATURE-----
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=WNej
-----END PGP SIGNATURE-----