Hacker News new | ask | show | jobs
by jedsmith 5627 days ago
> Oh, and reiserfs sucks.

I was with you until here. T'so is quoted as writing that btrfs has a lot of the same design ideas of ReiserFS, and that he supports btrfs as the way forward (he particularly disliked Hans and Namesys, though): http://lkml.org/lkml/2008/8/1/217

Some distros moved to ReiserFS by default, too.

Somehow I've never found the time to properly try it out, but I never really got the impression from others that used it that it sucks. I might be wrong, though.
2 comments
triffidhunter 5627 days ago
> I was with you until here. T'so is quoted as writing that btrfs has a lot of the same design ideas of ReiserFS,

When I was using reiserfs years ago, it ate all my data (/home unraveled). I like the features, and it was nice and fast for the time but I've held a grudge against it for years as a result. Looking at dumps of the on disk data structures showed massive corruption.

At work, we used it for a long time in a project where tail-packing is a huge win (many many small files). Never had any problems there.

So, it doesn't really suck I just had a bad experience.

link
wazoox 5626 days ago
I had too, at a time, and it was because of a flakey ATA controller (some VIA-based shitty mobo). Don't throw the baby with the bath water :)
link
kinetik 5627 days ago
Some distros moved to ReiserFS by default, too.

SuSE, who later realized it was a mistake and moved back to ext3?

link
jedsmith 5627 days ago
Even though they denied it, I'm willing to bet that SuSE (and others) made the move after realizing the Reiser community was about to be abruptly beheaded.
link
SoftwareMaven 5627 days ago
Not just beheaded (a particularly gruesome term in light of this topic), but to have a very negative association. Remember that SuSe is open-source but supported by the very corporate Novel. I don't know that I'd want that association with my company, and the suits running Novel are probably less inclined.
link
flomo 5627 days ago
The arrest was likely the instigator, but at that point it had long been clear that Reiser4 wasn't going to be accepted in the kernel mainline, and Reiser3 had gone off into an 'unsupported' limbo. The switch would have happened sooner or later.
link