On Mon, 7 Oct 2024 at 22:09, Yu, Max <[email protected]> wrote:
> Corruptions are unavoidable part of our life, disks and network can inject 
> failures due to unpredictable and unknown reasons like 
> https://www.sciencedirect.com/science/article/abs/pii/S0026271421003723. Even 
> multi-layer protection is not perfect as it depends on where the error is 
> injected.

I'm sorry but I am not convinced. You are the only one reporting the
issue, so I don't think it's unavoidable.

> We proposed this patch as it is aligned with how we currently handle 
> incorrect checksums: 
> https://github.com/yoctoproject/poky/commit/672c07de4a96eb67eaafba0873eced44ec9ae1a6.

What the patch is doing is essentially -c cleansstate on a cache used
by several consumers. This makes it a non-starter: you can't remove
items from live sstate like that. Once an object is in the cache, it
needs to be removed offline, or replaced atomically if it's corrupted.

Before we decide how to handle corrupted items, it's perhaps better to
consider first how they should be reported. Then users can decide what
they want to do with that information without yocto forcing something
on them. So please: how can I observe the issue?

Alex
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#205285): 
https://lists.openembedded.org/g/openembedded-core/message/205285
Mute This Topic: https://lists.openembedded.org/mt/108828269/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to