[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#991587: hw-detect: check-missing-firmware can be a tad aggressive

Control: clone -1 -2
Control: reassign -1 cdrom-detect
Control: retitle -1 cdrom-detect: ask hw-detect to skip missing firmware detection
Control: reassign -2 iso-scan
Control: retitle -2 iso-scan: maybe ask hw-detect to skip missing firmware detection

Cyril Brulebois <kibi@debian.org> (2021-07-28):
> While I'm not entirely certain about use-cases around iso-scan, I would
> think cdrom-detect should be one of the very few first things to happen
> on a system, and that raising errors because of missing firmware files
> at this stage wouldn't help: even a firmware-enabled image wouldn't be
> mounted and firmware packages wouldn't be available (later, they show up
> at /cdrom/firmware).
> Therefore, I'm proposing:
>  1. Implement support for an environment variable in hw-detect, that
>     would disable the check-missing-firmware call.

Done, but only tested manually (be re-typing the patch live).

>  2. Set this variable for both hw-detect calls in cdrom-detect's
>     postinst. [This should fix the problem I'm seeing.]

Done, will be tracked by -1 (same proviso as above).

>  3. Set this variable for the hw-detect call in iso-scan's postinst.
>     [Optional, I'm not sure what people are doing before iso-scan kicks
>     in; maybe some firmware-holding block devices could be available at
>     this stage, so letting the missing firmware detection in place might
>     make sense in this case.]

Not done, we'll keep track of it via -2 and see if people hit the issue,
mention their use case, and if we can draw any conclusions as to whether
we should go the same route as with cdrom-detect.

Cyril Brulebois (kibi@debian.org)            <https://debamax.com/>
D-I release manager -- Release team member -- Freelance Consultant

Attachment: signature.asc
Description: PGP signature

Reply to: