What is Reiserfs all about

[a_cute_bomber]

While Installing linux i saw a file system reiserfs
WHy is it given there at all when ext2 and ext3
work fine???

[tonybradley]

ReiserFS is a significant improvement over EXT2 and EXT3.

I'm not an expert, but as I understand it ReiserFS uses fast journaling to keep a running log of sorts of transactions occurring on the hard drive. Should the system lose power or need to be rebooted ReiserFS allows it to bounce back much faster without having to do extensive checks to validate the drives before mounting.

When a system is started it fsck checks the meta-data to make sure the filesystem is in a useable state. If it finds that the volume was not properly unmounted it assumes that the meta-data is inaccurate and begins an exhaustive scan against the meta-data to correct any errors. If you have a large hard drive this can take a LONG time- sometimes an hour or more.

By using journaling ReiserFS maintains a running database of transactions with the file system. In the event of a failure or improper shutdown, fsck looks at the journal and only scans the portions of the meta-data that were changed recently. Using this method the fsck scan can be completed in seconds instead of hours.

It is also more space efficient. In EXT2 a file will use 1024 bytes (1 kb) of space even if the file is only 80 bytes. ReiserFS can write blocks of any size so the 80 byte file will only occupy 80 bytes of space on the drive. Multiplied by hundreds or thousands of files this can save a LOT of disk space.

I don't know what new features they intend to add, but I believe ResierFS version 4 is set to come out in June of this year.

[problemchild]

What he said.

And I would add just a couple of things..... ReiserFS was the very first journaling file system for Linux, having been available as a kernel patch for about 3 years now. Mandrake and SuSE were the first to adopt ReiserFS because they were known as bleeding-edge distrubutions and they knew a lot of people wanted a journaling FS. However, ReiserFS quickly proved to be a mixed bag because as good as it is, it uses a radically different B-tree structure than other more conventional filesystems, and that has caused a number of incompatibilities with other parts of the kernel. Those problems can and have caused data corruption and data loss under certain circumstances. ReiserFS has a very checkered track record, and a lot of people (including me) don't trust it.

I agree with Tony's statement that it represents a huge improvement over ext2, but I disagree that it has anything over ext3 for most purposes. Ext3 has the same journaling capabilities as Reiser, with the additional feature that it can optionally journal the actual data on the drive (not just the metadata) at an amazingly small performance penalty. That means your data is safer. Although a lot of people claim that ReiserFS is extremely fast, I haven't found it to be significantly faster than ext3 for most purposes. ReiserFS can be an advantage in specialized situations where you have lots and lots of very small files, but otherwise it offers me nothing over ext3, and why risk a file system with a history of data loss?

So I would be hard pressed to recommend ReiserFS as an everyday file system. All things being equal, I think ext3 is a far better bet for most people.

[Louie]

I didn't like ext3. It took longer to recover from reboots then ReiserFS. I think with current linux kernels and the new programs there are very limited occurences of data loss with ReiserFS.

Heck if my grandma can run ReiserFS so can I!

[gore]

this is the ReiserFS thing i typed out in the redhat 9 Post, ill copy and paste it for you here because i think i did a good job typing it out, it should help

Well, i am a SuSe person and i use Reiser FS so ill give some details for you all (and you better appreciate this because its a hell of a lot of typing) anyway, heres some specs:

REISERFS
Officially one of the key features of the 2.4 Kernel release, Reiser FS has been available as a kernel patch for 2.2.x SuSe kernels since SuSe Linux version 6.4. ReiserFS was designed by Hans Reiser and the namesys developmentteam. Reiser FS has proven to be a powerful alternative to the old Ext2. its key assests are better disk spave utilization, better disk access performance, and faster crash recovery, the only real drawback is that it pays alot of detail to metadata but not the data itself. Future versinos will include data journaling, both meta data and actual data, are written to the journal, as well as ordered writes.

Fast crash recovery:
using a journal to keep track of recent metadata changes makes a file system check a matter of seconds, even for huge file systems.

Better disk access performance:
for small files, you will often find that both file data and "sts_data" (inode) information are stored next to each other. they can be read using a single disk IO operation, meaning, that only one access to disk is required to retreive all the information needed.

Better disk space utilization:

in Reiser FS, all data, is organized in a structure called B*-balanced tree. the tree structure contributes in better disk space utilization as small files can be stored directly in the B*- tree leaf nodes instead of being stored elsewhere and just maintaining a pointer to the actual disk location. in addition to that, the storage is not allocated in chunks of 1 or 4 KBs, but in portions of the exact size that is needed. Another benefit lies in the dynamic allocation nodes. this keeps the FS more flexible, in contrast to traditional file systems, like Ext2 where the inode density has to be specified at file system creation time.


More specs:

Reiser FS:

File size Limit: 1 EB

File system size Limit: 16,384 GBs (16 TB)