Re: Converting RAID1 to single fails if RAID1 is missing device

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

在 2025/2/3 15:52, Colin S 写道:
> On 02/02/2025 20:32, Qu Wenruo wrote:
>>
>> But I think the ultimate solution is to make btrfs to properly detect
>> and support device shut down request.
>
> Yes. How many years and how many more users need to have this problem
> before it’s given some priority?

Complaining is so easy that some one doesn't even know what's going
wrong can do.

>
>> Although that would also introduce new complexity, e.g. what if the
>> missing devices show up again after missing several writes?
>
> Since I see you wrote a patch in 2022 to add write-intent bitmap for
> raid5/6, don’t you already understand the answer is a write-intent
> bitmap? Further, did you not see any of the several messages I have sent
> to the mailing list talking about exactly this in the last year? I am
> genuinely confused.
>


It's completely a different bug.

It doesn't even have RAID56 involved.

And write-intent bitmap is not the solution at all. It doesn't support
things like zoned device support (which is now part of the core btrfs
functionality).

Knowing what you don't know is important.