Draft bootloader selection test cases

Adam Williamson awilliam at redhat.com
Sat Aug 3 01:20:49 UTC 2013


On Wed, 2013-07-31 at 04:08 -0400, Kamil Paral wrote:
> > I drafted up a couple of test cases to verify this currently
> un-verified
> > criterion:
> > 
> >
> https://fedoraproject.org/wiki/Fedora_20_Final_Release_Criteria#Bootloader_disk_selection
> > 
> > They are:
> > 
> >
> https://fedoraproject.org/wiki/User:Adamwill/Draft_QA_Testcase_anaconda_bootloader_target
> 
> " Manual verification with dd, efibootmgr etc. should confirm
> installation of the bootloader code to the selected disk "
> 
> Not that many people will know what this means. Can you add sample
> commands how to do this? How do I tell that bootloader is present
> using dd?
> 
> >
> https://fedoraproject.org/wiki/User:Adamwill/Draft_QA_Testcase_anaconda_no_bootloader
> 
> " Boot the system from another bootloader (if you used a test system
> with a working bootloader configuration) or a live image and verify
> using dd, ls, efibootmgr etc. that the installer did not install a
> bootloader "
> 
> The same comment as above, and more. This is actually even harder. If
> we want to compare the first xxx KB of the disk, we need to make a dd
> copy in advance, before the installation. So that should be mentioned
> in the test case, otherwise people find out too late.
> 
> "Boot the system from another bootloader" means that I should try to
> boot the newly installed system from a bootloader placed on a
> different disk (in this case, configuring all the correct parameters
> is really hard and just a few people will be able to do it without
> instructions), or that I should boot a completely different OS
> (already present) and use dd for verification that boot area wasn't
> touched. 

Thanks for the feedback. I didn't want to get into too much detail in
the test case, because things can be fairly different on different
arches, BIOS vs. UEFI etc. But it might be a good idea to include some
examples for the most common cases, I guess.

> In the latter case we will verify that the bootloader area wasn't
> touched, but we won't verify that the newly installed system boots.
> Which is... suboptimal. The same goes for LiveCD verification.

I'm not sure that actually _is_ suboptimal, because if you're not
installing a bootloader, there's really nothing at all the installer
does to render the installation 'bootable'; there's nothing to test. It
just installs the packages. It's up to the user to configure boot. So
I'm not sure there's something it makes any sense to test, at that
point.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net



More information about the test mailing list