Bcachefs Linux File-System Seeing Performance Improvements, Other Progress

[pal666]

Yeah? Maybe cases don't reach actual bugzilla because users give up before devs get around to address their case year-two later.
https://forums.opensuse.org/showthread.php/516753-Corrupted-BTRFS

maybe you shouldn't compare unwritten piece of shit to filesystem 4 years ago?
and btw your link is funny, it has nothing to do with btrfs, just some clueless user. was it you?

[pal666]

i like it claims. but i would verify them as soon as it hits mainlain

if i double all his claims, will you pay me?

[pal666]

Can you please stop pretending to not understand how announcements about future milestones work?

i don't understand imbeciles who judge software based on announcements about future milestones. everything works in powerpoint

[flower]

do you like bcachefs, or claims of it author? and current one or future fantasy?

i like it claims. but i would verify them as soon as it hits mainlain

[pal666]

Not seeing how making claims about intents is bad.

the bad thing happens when average imbecile thinks that others don't have such intents

You want to see the original claims for btrfs? It was supposed to be much cooler than it probably will ever be, as someone grossly underestimated the complexity of dealing with stripes (raid5/6 mode).

i.e. all you can come up with against btrfs is raid56? ok. btrfs raid56 was copied from block level raid and inherited its stability issues. so btrfs doesn't have raid56 atm. but no other fs has it anyway, so there's no competition. it isn't fixed yet because there's no demand for it(not forum poster demand, demand with money to pay for development). i'm sure it will be fixed eventually, right now use raid1, it will be faster as a bonus point

[pal666]

Nonsense. FS that won't eat your data" clearly relates to future point of introduction.

so who faster will reach future point of "FS that won't eat your data" - one guy on donations, or dozen of full-time developers?

You shouldn't expect that from a product that hasn't made its way into official kernel .

and you shouldn't be excited about something which doesn't exist

For what it is, it looks GREAT. I like it.

what it? website? unsubstantiated claims about future? source code with botched rebases?

If I weren't unemployed ATM, I'd definitely send a few € his way.

and if you were less gullible, you wouldn't

As with any great idea, this FS is obviously facing some resistance on its way, but I don't doubt it has a great potential.

there's no resistance. fs doesn't exist(you said that). there's some guy hyping clueless people for donations. current in-tree filesystems have much greater potential

It checks so many boxes on my wishlist and so many options that are not likely to be found on ext4 & Co. in foreseeble future.

sadly your wishlist is missing "no bullshit" check box

It's obvious that this is fruit of the real engineering and experience and not merely a programing masturbation.

are you even a programmer? did you look at sources?

[starshipeleven]

lol, are those stability claims for current state or for some imaginary future version? then every filesystem is rock-solid in some far bright future

Not seeing how making claims about intents is bad. You want to see the original claims for btrfs? It was supposed to be much cooler than it probably will ever be, as someone grossly underestimated the complexity of dealing with stripes (raid5/6 mode).

[starshipeleven]

This is really interesting.

Either I believe JustinTurdeau or starshipeleven and think that btrfs has never had any issues with anyone ever and it's only ever used error.

Or I believe duby229 and think that btrfs can still have issues that corrupts data.

That's not my argument. I'm just disproving his bullshit claim that "btrfs always corrupts itself when doing a balance".

If your system does that with CURRENT btrfs, you have a hardware issue, period.

I'm assuming duby229 is using something else which is stable for him.

Probably Windows.

In my case, I've only run into three filesystem issues. JFS on old hardware, EXT4 corruption bug, and BTRFS.

The difference is that the EXT4 corruption bug didn't destroy all my data, and JFS didn't destroy any data, it just slowed down really bad.

I started using btrfs in 2018 when I migrated a couple PCs and a server to OpenSUSE (first class btrfs support was one of the main reasons I chose OpenSUSE), and so far I've not seen any horror story, and it has survived some pretty intense abuse by me, yanking the power from the PSU that powers the drives in the array, or forcing power off when the PC locks up (OOM issues, unrelated to btrfs), disconnecting drives by mistake while it's operating while I'm moving other hardware in the case....
I run a full scrub, balance and dedup weekly (where applicable anyway, not all systems have it in RAID mode).
All drives/arrays use zstd compression algorithm.
Never used RAID5/6 though.

[pal666]

That said i actually like bcachefs and would like to use it.

do you like bcachefs, or claims of it author? and current one or future fantasy?

[pal666]

First, the reliability track record was about bcache, not bcachefs.

i.e. it was used to mislead users and backers?

Second. The filesystem is under development, pre-release, v0.0.0, whatever else do you call it. The "won't eat your data" part relates to its architecture and development model — which, by author's intention, will translate to better reliability once the major groundlaying development is done and the filesystem is actually released.

what are you trying to imply, that every other filesystem has architecture and development model aimed to eat users data? that's same bullshit misleading users and backers. all filesystems have intention to not eat data. but people make mistakes from time to time. i.e. all bcachefs claims are load of bullshit