Zygo

If you get an inode with `ls -li`, can you look up its inode with `btrfs ins ino .` on the 64-bit kernel? If not, you may have found a...

4 GB per hour is 35TB per year, which takes decades to wear out a modern SSD [1] unless it's a very old or low-budget model. It should stop writing...

The calculation tells you when the drive reaches the end of its TBW warranty (100% of endurance used, which is a complicated function of the lifetime of several different flash...

Samsung PRO models do the job, but they are the most expensive consumer drives for what they do (you _can_ pay more, but those drives are specialized for video production...

It looks like you have a large item in the root tree metadata? Can you provide the output of: `btrfs ins dump-tree -t 1 /dev/your-btrfs-device-here` The code was rewritten between...

We're looking for something with a large 'itemsize', especially if it's > 4096.

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 many...

You do have a point about write multiplication, though: if the minimum hash-extent (we really need to rename those) size is 128KB, and every 4K of new data triggers at...

bees only reads the hash table once, then writes dirty pages as they are dirtied. So hash_extent_in is going to be the number of hash table extents (unless there are...

BEESNOTE sets up a callback to report the current thread status. The BEESNOTE text is incorporated in a closure stored on a stack for each thread. When bees does print_status()...