Kai Krakow

@charles-dyfis-net Is it compatible with existing configuration files? Read: Is there an upgrade path without bad surprises?

@charles-dyfis-net Thanks, that would make reviewing much more motivating :-) Because I'm open for porting the good ideas of it over here.

@charles-dyfis-net Give me few days to test this new wrapper here locally with Gentoo. I may create a PR with some simple tweaks so it could end up in a...

What does a Nix config file look like?

Ah okay, I start to understand. So this is declaration-based config generator what we see here. Why does it need a complete rewrite of the service wrapper then? Why not...

These are a lot of valid points, some of them I also raised here already (tho, with other words).

Here's another system showing the same behavior:  And another which doesn't show this behavior:  The last system doesn't have the rootfs on btrfs, only a storage pool for...

Maybe there's a feedback loop between bees logging to journal, and then bees seeing new extents and flushing seemingly dirty hashes? Although journal writes to a nocow/nodatasum directory...

> The hash table should be tracking which hash extents are used and only writing out those that were touched. That said, extents are pretty big, and it doesn't take...

Maybe this is an effect of resizing the hash table? The affected machines both had their hash table size doubled without resetting `$BEESHOME`. There's this code before the cells loop:...