Re: [REGRESSION] ext4: silent data loss when a split at I/O completion gets ENOSPC

From: Jose Fernandez (Anthropic)

Date: Wed Nov 18 2026 - 09:06:29 EST


On Wed, Nov 18, 2026 Zhang Yi wrote:
> Could you provide a reproducer? My guess is that high concurrency causes
> the blocks freed by my current process to be taken by I/O issued by
> other processes

That fits our runs. I'll send you the script off-list.

- Jose