To what extent has ReiserFS's popularity been lessened by Hans Reiser?

Question

How does the future look for ReiserFS? Did the trial affect its popularity much?

This is a pretty subjective question, so referring to a published statistic/article would be nice.

Answer 1

reiserfs is currently maintained upstream in the kernel. I found 4 changes in the 2.6.30 release notes. There are also efforts (which led to some of the 2.6.30 commits) to reduce the use of the Big Kernel Lock (BKL). The use of the BKL is a major bottleneck for reiserfs on Multicore systems and should show some performance gains by being nable to lock on just the superblock instead. Novell & SuSE have continued supporting the file system for their enterprise product so it's not dead yet. Also, it's often not 'officially' supported by many of the distros, but you can for instance install the latest Fedora release on ReiserFS (with installer boot params linux selinux=0 reiserfs).

As far as Reiser4, it is currently in Andrew Morton's -mm tree and development continues led by Edward Shishkin (a former Namesys employee). In a mid-May email on reiserfs-devel, Edward estimated that the requisite changes for merging it into Linus' tree of the kernel might get done in the August/September time frame.

People seem to have moved on to getting excited about ext4/btrfs/ZFS, so reiser4 going into the mainline may be more of a victory for its supporters than anything else, but at the same time if it's better than the other available file systems then I'm sure it will regain popularity.

Answer 2

The Wikipedia article says that development has stopped on ReiserFS v4, so I would take that to mean that you're at the mercy of your host OS's support for ReiserFS v3 and the year 2038, whichever comes first.

You're right, it is a very subjective question. Especially if you're trying to gauge its popularity (which I think is a pointless exercise).

Answer 3

If you need to ask, you know the answer.

Answer 4

His severely poor social skills caused him to be unable to work with the kernel developers. For example he had good ideas that required API changes. A compelling case must be made for these, and he refused to make his case. He often did not listen to other people of similar intelligence, and refused to make technical arguments to an audience that requires them.

When reading details of his trial, it seems that these problems in socialization are related to the issues that led to the tragic events of his life. But I don't believe that the trial would have had such an impact on the filesystem's popularity if he had been more willing to work with others in a constructive manner.

Answer 5

This LWN article from 2008 is a decent description of the context.

Mr. Reiser's role in reiserfs development and maintenance ended some years ago, though. He stopped work on it when reiser4 development started, and even opposed the incorporation of improvements done by others. Reiserfs continues to be maintained independently of its creator, though there is not much interest in adding features to it at this point. Reiserfs is nearing the end of its run, and nothing which happened this week has changed that situation in any way.

Answer 6

Work still goes on - see this list:- http://www.nabble.com/ReiserFS---General-f1023.html

Answer 7

Well, ReiserFS is a nice reasonably bug free file system in the kernel. Reiser4 seems to be a dead end, and not likely to ever make it into the kernel.

For those of us interested in the idea of the Reiser file systems, and also of ZFS, the fs to watch is BtrFs. Still alpha though.

http://btrfs.wiki.kernel.org/index.php/Main_Page

Answer 8

Better questions would be

"Is there being done any maintenance work on the current Reiserfs? (v3)" "What happened to Reiser4? It was very promising"

I was big fan of Reiserfs and I also tried some pre-alpha of Reiser4 that was quite fast as well as extendable. Would love a filesystem that can store meta information, such as ID3 tags.

Answer 9

Not for me. I benchmarked reiser3 against ext2, ext3, and XFS. I didn't have difficult requirements, simple journaling and low overhead was what I was looking for. Ext3 beat the others hands down in speed tests. At the time, my thought was "if it isn't faster, and I don't need the extra features, I don't get what the fuss is about."

I already disliked it, I guess is what I'm saying.