On Thu Apr 17 10:11:32 2025 +0000, Nikolay Sivov wrote:
It's possible LVM_ARRANGE is what's causing REORDER event. But I'm not sure how deep we want to go with this, there are too many paths.
Could be. I could see doing that if we have enough test coverage to actually work out those details. For now, I don't think it really matters, as long as we send the events we're supposed to, in the order we're supposed to, and only at times when the control is in a consistent state with the change already in effect.