BcacheFS vs. EXT4 vs. Btrfs vs. XFS vs. F2FS

[macemoneta]

Never had any silent corruption with ext4.

That's the thing about silent corruptions; you don't know they are happening until you go to get a file and it has been trashed. You think everything is fine, but it's not. Unless you run an 'rsync -cv' to validate the backup, you won't know that the backup is invalid.

[_ck_]

Is it possible to start measuring latency (or at least max latency) with the test suites?

Because EXT4 really seems to shine on latency when all other measurements between FS are too close to call.

[renkin]

Why isn't the autodefrag option ever used in the btrfs tests? I really want to know how much of a difference it actually makes.

[bug77]

Why isn't the autodefrag option ever used in the btrfs tests? I really want to know how much of a difference it actually makes.

On an SSD? Lol.

Also, I'd like to that I like ext4 because I can still use the old ext2 Windows driver to read my Linux partitions.

[carewolf]

I'd rather stick to btrfs; I've had too many silent corruptions with ext4.

I prefer a silent corruption to a filesystem that bricks the first time it runs into a corruption.. Good job detecting, now mount you piece of shit!!!

[Stellarwind]

I prefer a silent corruption to a filesystem that bricks the first time it runs into a corruption.. Good job detecting, now mount you piece of shit!!!

It depends... if it is just your torrents, then you don't care and ext2 will be ok too. If it is something like your corporate accounting data, then you probably want it to detect and fail right away.

[CrystalGamma]

It depends... if it is just your torrents, then you don't care and ext2 will be ok too. If it is something like your corporate accounting data, then you probably want it to detect and fail right away.

My guess is that carewolf wanted to complain about a corrupted FS being hard to recover because it blocks on mount, not say that silent failure is a good thing.