Novell Moves Away From ReiserFS

VSquared56 writes, "Novell announced a shift in the default filesystem from ReiserFS to ext3 for users of its SuSE Enterprise Linux. This news comes shortly after Hans Reiser's arrest, though Novell says the decision was being considered long before. Though Novell will continue supporting ReiserFS 3, it claims ext3 is more stable and will 'soon' match performance with the newer ReiserFS 4. What implications will this have for SuSE users, and ReiserFS's future as a whole?"

Comments

It's Deja Vu All over Again

[joe_n_bloe]

I bet we see this article once every day for the next ... well, till the trial is over.

Re:It's Deja Vu All over Again

[arth1]

I predicted this a few days ago, when I wrote here:

That it /is/ going to damage reiserfs is beyond any doubt, no matter whether he's proven innocent, not proven guilty, or proven guilty. The name is tainted, and a business executive will not likely touch anything related to that person, no matter whether it gets taken over and run by other people or not.

This was modded flamebait.

People, you might not want to hear it, and you might not agree with stupid knee-jerk reactions, but these reactions will be coming. The name "reiserfs" is tainted, whether that's rational or not.

Regards,
--
*Art

Re:It's Deja Vu All over Again

[bernywork]

Unfortunately, I hate to agree with you, but it's true. If they project renames and then it continues, it might get picked up again.

The other concern is going to be about support, if Hans is found guilty or not, it doesn't really matter. A company such as Novell may consider that the filesystem platform isn't as supported as what it once was and is moving away from it.

From a marketing point of view, Novell won't want to associated with it either. If they show support for him, and he is found guilty, it's a marketing nightmare for Novell.

Re:It's Deja Vu All over Again

[Professor_UNIX]

A company such as Novell may consider that the filesystem platform isn't as supported as what it once was and is moving away from it.

Good. They should be throwing their weight behind a better filesystem like XFS or JFS instead. If you're dealing with large files a lot, they put ext3fs AND reiserfs to shame. The only place I'd use ext3 anymore is on my system disk, but my large video stores run JFS.

Re:It's Deja Vu All over Again

[donscarletti]

If you're dealing with large files a lot, they put ext3fs AND reiserfs to shame. The only place I'd use ext3 anymore is on my system disk, but my large video stores run JFS.

Yes, JFS and XFS suit you better than Reiser4, but not everyone USES big files, everyone uses small ones though. [XJ]FS suits big files better, Reiser4 supports little ones better, it doesn't mean that one is better than the other.

Re:It's Deja Vu All over Again

[Anonymous Coward]

The name "reiserfs" is tainted, whether that's rational or not.

Apparently it is to be called "icefs" in Etch.

Something to do with Hans not being available to QA patches by the Debian kernel team.

Re:It's Deja Vu All over Again

[tomhudson]

From a marketing point of view, Novell won't want to associated with it either. If they show support for him, and he is found guilty, it's a marketing nightmare for Novell.

Are you kidding - this stuff practically writes itself.

1. "Novell and ReiserFS - we have THE killer linux filesystem!"
2. "Novell and ReiserFS - your files are as secure as the county lockup."
3. "Novell and ReiserFS - while the jury's out, we continue give you a choice of file systems"
4. "Novell and ReiserFS - exabytes of data - and only one piece of information missing ..."
5. "Novell and ReiserFS - no dead inodes, no dead files, no dead bodies ..."
6. "Novell and ReiserFS - what the fsck?"

Re:It's Deja Vu All over Again

[Jeremi]

People, you might not want to hear it, and you might not agree with stupid knee-jerk reactions, but these reactions will be coming. The name "reiserfs" is tainted, whether that's rational or not.

Eh, that's easy enough to overcome. Namesys will just release the next version of ReiserFS as "ext5".

Re:It's Deja Vu All over Again

[Anonymous Coward]

Reiser4 supports little ones better

Can you give an example or quote a source or something? It may support smaller files better than XFS, but JFS is an enterprise quality filesystem, and is better overall than both Ext3 and ReiserFS, and no, Reiser4 doesn't do anything better than JFS yet.

If IBM hadn't wrote it, most distros would use JFS.

Re:It's Deja Vu All over Again

[kimvette]

What would happen if (god forbid) Linus were in Reiser's place? Would everyone here be distancing themselves from using Linux? Would Novell, IBM, etc. abandon the use of Linux, throwing out the baby with the bathwater?

This is why when the story hit I posed legitimate questions regarding the filesystem's future (and got flamed for it, BTW, here and on linuxquestions); a person's career work should be viewed independently of his or her personal misdeeds. Otherwise, we should abandon electricity and incandescent lights (Edison was a bit of a bastard, and his invention of the electric chair "tainted" AC), jets (Heinkel was a nazi), Mercury and Apollo programs should never have happened (Wernher von Braun, the brain behind those programs, was a nazi, willing or otherwise). There are many, many worthwhile inventions proposed, designed, and/or implemented by evil people, and yet we use them on a daily basis, because regardless of the creators' nature, philosophy, or misdeeds, they have produced some worthwhile things that abandoning them because of the heritage would be somewhere between silly and irresponsible.

Re:It's Deja Vu All over Again

[Anonymous Coward]

"If IBM hadn't wrote it"

That's writted, fucktard.

Re:It's Deja Vu All over Again

[Anonymous Coward]

You missed:

"Novell and ReiserFS - Once you're with Reiser, you'll never want to leave."

I almost posted that joke hours ago, but was too ashamed to until I read your post. Thanks for lowering the bar for us all.

You know, what, bar's still not quite low enough. Let's make sure that "Post Anonymously" is checked...

Re:It's Deja Vu All over Again

[carrett]

The difference between Reiserfs and Linux is that Linux has contributions from tons of developers worldwide. In fact, though Linus is the BDFL, I don't think he writes huge portions of the code (anymore). If this happened to Linus, Linux would just need a new leader (first mate gets promoted) and maybe a name change (though I bet there are more people who know about Linux than there are people who know who Linus is, so maybe not), whereas now that Hans is in deep doo-doo, there's no obvious person to take his place.

I don't think we should lose any of the innovations Hans contributed either, and I hope someone worthy comes along to continue his work, but I don't know if Hans ever really had a "second in command."

Re:It's Deja Vu All over Again

[Anonymous Coward]

Friday afternoon, a good friend with whom I wished to become even closer came to visit, to have Linux installed on his vintage PC.

No DVD drive, that much was known, so I had dutifully prepared a SuSE (Novell) boot CD for the installation.

The plan was to boot from that CD, and then do the installation via the network (the plan was to connect his computer to mine using a gay cable).

No luck, his computer had SCSI CD drives, which its BIOS was unable to boot from :-(

Next try: boot from floppy.

Unfortunately good floppy disks are hard to come by, and I had to go DEEP into my closet to find some. Predictably, while digging, all kinds of other stuff fell out: lube, condoms, and a book "Outing Yourself -- How to come out as a lesbian or gay to your family, friends, and coworkers". That was not a big problem, as the good friend was on the bus too.

A bigger problem was that among all those floppies, most have gone bad during all these years...

Eventually, I could dig up enough good floppies to make SuSE boot floppies and boot my good friend's PC from them.

Next surprise: the network card was an ISA card. Arghhh! No way to autodetect that!

Ok, so next try with SuSE 9.3 (of which I had CDs) rather than 10.1 (for which I only had DVD).

All went fine until it started formatting the disk: bad sectors (did I tell you that this was old refurbished hardware). No panic: the errors were near the end of the disk, so I just shrunk the partition, and did a mkfs.re... no, no, I can't type the name of a murderer. So I typed mkfs.ext3 manually, mounted the partition, and clicked "ignore error" into the install menu.

Install went mostly smoothly, until reboot. "Cannot mount root filesystem". Oops.

Yes, of course, SuSE was not really expecting ext3, it was not even included into its initrd. In the meantime, it started to become late, and I was having another meeting in the evening. Ok, so I had to shove in the 8 floppies to boot the rescue system, and include ext3. Fortunately, despite their name, floppies are still stiff enough to enter the hole, hehe.

Still this filesystem episode ate about half an hour of our time, enough that by now it was too late to do anything other than Linux. Bummer! It could have been such a hot afternoon, were it not for that wretched filesystem!

Re:It's Deja Vu All over Again

[Anonymous Coward]

What would happen if (god forbid) Linus were in Reiser's place?

Given that he is married, this is a definate risk.

In order to be really protected against such issues, the lead developers need to stay celibate. And better be gay!

Re:It's Deja Vu All over Again

[neonmagic]

Hans has been unstable for quite some time, refusing to abide by Kernel coding requirements, having childish arguments on the LKML and outbursts etc. He's never quite struck me as a sane person in all honesty, but I would not have painted him as a killer either, simply a weird geek with very poor social skills.

Hans is the lead programmer for ReiserFS, he sets to the directions that it'll head in. Without him at the helm, the project will lose focus.

As to whether he's guilty, that's for a judge or jury to decide, but the evidence I've seen reported so far isn't very convincing of his innocence in all honesty. Blood in his house and car? Missing rear car seat? Not co-operating with police on the location of said car? He was the last to see her alive? Books on police homicide investigation techniques in his house? Rather damning evidence in all honesty.

I feel very sorry for his daughter.

Dave

Re:It's Deja Vu All over Again

[twistedcubic]

The last time I used XFS was about 5 years ago. I remember it took 50 hours to delete a 500MB file. You say XFS is better than EXT3 for large files. I assume things have changed, then, no?

Re:It's Deja Vu All over Again

[TrentC]

What would happen if (god forbid) Linus were in Reiser's place? Would everyone here be distancing themselves from using Linux?

This is one Linux distributor making a decision for their product; a decision which may have been in the works even before Hans' arrest, though I assume that may have made the decision an easy one.

Despite my opinion that Hans is guilty, I don't hold it against the software he created. If the remaining developers want to rename Reiser4 to avoid the ugly association, fine.

And, no, I don't think "if you aren't going to use ReiserFS because Hans killed his wife, then you should stop using electricity because Edison created the electric chair" is a legitimate question; it's a straw argument meant to elicit a specific emotional reaction. (For one, I'm pretty sure electricity existed before Thomas Edison did.)

Re:It's Deja Vu All over Again

[Corwn of Amber]

Better? I never, ever found anything faster than ReiserFS. I benched ext3 vs ReiserFS 3.6.. Guess who won...

I'll have to try JFS and XFS some day.

Re:It's Deja Vu All over Again

[GoRK]

ReiserFS when questioned stated, "I've deliberately cut your files into bits and hidden them in every last available nook and cranny. Since you were relying on me completely to remember where I put everything, I found it quite convenient to simply forget, leaving you with no way to make sense of the leftover mess. Good luck pinning anything on me!"

Re:It's Deja Vu All over Again

[ajs318]

Edison did not invent electricity. He re- invented the filament light bulb (which had already been invented a year or so previously by Joseph Swan), which turns most of the energy fed to it into heat (and only a very tiny portion into visible light). Given that almost every other conceivable solution for turning electricity into visible light is better, the filament light bulb is an invention which really should be consigned to the dustbin of history as soon as possible.

But that's nothing to do with what Edison was like. It's simply because his invention is, by modern standards, crap. Shame on anyone who is still using filament bulbs.

Re:It's Deja Vu All over Again

[prattle]

Something to do with Hans not being available to QA patches by the Debian kernel team.

I thought it was because Reiser is expected to become non-free.

Re:It's Deja Vu All over Again

[greenrd]

He continued "Hey, haven't you ever heard of backups? Maybe you should keep regular backups next time."

Re:It's Deja Vu All over Again

[Anonymous Coward]

I hope you've torn the dash out of your car to replace incandescents with LEDs, ditto for marker lamps, courtesy lights, etc. as well as upgraded to an HID system if your car did not come from the factory with it.

Have a mag light at home (what geek doesn't have two or three)? Have you upgraded it with LEDs and put up with decreased output? No? Hmm. Shame on you!

Re:It's Deja Vu All over Again

[ajs318]

What makes you think I have a car?

Re:It's Deja Vu All over Again

[Christopher Cashell]

Things have changed significantly.

Deleting a large (hundreds of GB) file in JFS or XFS is nearly instantaneous. The same action on an ext3 filesystem can take up to a minute, even on a fast machine.

I still use ext3 for /boot and sometimes for / (purely because there are still more and better recovery tools for ext3 than any other Linux filesystem (due in part to it's simplicity, which is also the source of it's limitations)), but /var, /home, and all data storage drives are JFS or XFS.

Re:It's Deja Vu All over Again

[Anonymous Coward]

I thought it was because Reiser is expected to become non-free.

Hey, it's innocent until proven guilty remember. Maybe you know something we dont?

Let this be a lesson to you all

[nih]

Crime does not pay!

Re:Let this be a lesson to you all

[Anonymous Coward]

Crime does not pay!

Ok, ReiserFS might be a bit on the slow and unstable side, but I would not actually call it a crime.

Re:Let this be a lesson to you all

[bcat24]

Umm, whatever ever happened to innocent until proven guilty? I agree there may be a good chance Reiser did it, but I think there's also reasonable doubt he didn't.

Re:Let this be a lesson to you all

[Anonymous Coward]

Or innocent unless proven guilty. The "until" saying seems to imply that they are guilty and it is a matter of time until their guilt is proven.

Re:Let this be a lesson to you all

[TrentC]

Umm, whatever ever happened to innocent until proven guilty?

Ask José Padilla.

Snarkiness aside, the answer is simple; I'm not a member of the jury, nor am I the judge trying his case, so I have no legal obligation to assume that he is innocent until proven otherwise.

From what little I've read about the actual case, I believe he did it. If I read more that reasonably demonstrates that he did not do it -- as opposed to crackpot Slashdot theories that blood found in his car could have come from a nosebleed, or that Hans Reiser bought books on criminal investigation techniques to satisfy his curiosity -- then I will change my mind. If a jury of his peers finds him not guilty, then I will accept their verdict.

But I'm not required to think the best of Hans Reiser, and this "whatever happened to innocent until proven guilty?" meme only reinforces my belief that there are a bunch of shallow, self-indulgent people on Slashdot.

Hurm

[OverlordQ]

What implications will this have for SuSE users

Well, just a guess . . . but they might have to use a new filesystem!

Re:Hurm

[philwx]

Well, they had to make up a question (no matter how vague) to get this article on slashdot.

Re:Hurm

[gerrysteele]

Or, at a guess, will have to manually choose the filesystem they want when they install. I'm sure it will impact like zero people.

Have you ever heard the phrase:

[Anonymous Coward]

"Rats are the first to desert a sinking ship"?

Re:Have you ever heard the phrase:

[Anonymous Coward]

If the ship is truly sinking (it's probably too early to tell), then wouldn't you rather be the first to leave than the last? Or would you rather be a possibly dead hero?

Jeff

Batman

[Anonymous Coward]

Holy filesystem batman! :O

Re:Batman

[Anonymous Coward]

Was that supposed to be funny?

Re:Batman

[Anonymous Coward]

Well, it was posted anonymously, so maybe the author had doubts. Still, I haven't seen much batman, so maybe I don't get some reference.

The Slashdot Way...

[xquark]

I think is to have a poll as to measure people's opinions
about the guy's innocence. With options such as
1. He is innocent
2. He is guilty
3. Cowboyneal did it etc..

Re:The Slashdot Way...

[Anonymous Coward]

There's a logical explaination for everything

Re:The Slashdot Way...

[dvice_null]

I think anyone capable of writing a filesystem is capable of killing a person. But also anyone writing an open source software is too kind to kill a person.

Re:The Slashdot Way...

[WilliamSChips]

ESR? Oh wait, you mean actual programmers.

Re:The Slashdot Way...

[Wdomburg]

Erm, he wrote fetchmail, bogofilter, hexdump. The original sed distributed with the GNU utilities was written by ESR. He's got code in Gnome, Gnuplot, libpng, emacs, screen, etc.

Regardless of what you think about him personally, it's hard to dispute that he's an "actual programmer".

Re:The Slashdot Way...

[WilliamSChips]

Yes, I know the stuff he actually wrote. It's a joke based on the fact that he hasn't done much programming recently.

Re:The Slashdot Way...

[carlmenezes]

I think anyone capable of writing a filesystem is capable of killing a person. But also anyone writing an open source software is too kind to kill a person.

Exactly. Therefore, I believe temporary insanity is the only defense that applies.

xfs for ever

[eneville]

why not move to xfs? it's a very good performance file system. unless there a rumours of the author being a murderer of course.

Re:xfs for ever

[arth1]

why not move to xfs? it's a very good performance file system.

XFS is high performance especially for large files and multitasked access.
reiserfs (3) is high performance especially for small files and singletasked access.
JFS is also a good journalled file system with many nifty features, although perhaps not as mature as XFS.

unless there a rumours of the author being a murderer of course.

Neither X nor J have been accused of murder, to my knowledge.

All hail J.

Regards,
--
*Art

Re:xfs for ever

[ozmanjusri]

why not move to xfs?

Both XFS and EXT3 are more of a step sideways than a step up. I'd love to see a mainstream Linux distro adopt Sun's ZFS as its default filesystem.

Re:xfs for ever

[cortana]

There have been too many reports in the last couple of months of people whose machines have lost power, and booted up, only to find that every file on their XFS filesystems has been filled with zeroes.

Re:xfs for ever

[EsbenMoseHansen]

For the record, I was also quite underwhelmed by XFS. The Gentoo people, I think, wrote that XFS is primarily to large files and *only* if you have an UPS (and proper shutdown control). The problem is that it (quite aggressively) cache write-data; I have seen data disappear which was written nearly 2 hours before. I am quite happy with ext3. Reiserfs had a nasty tendency to slowly deteriorate over time, becoming slower and slower.

Re:xfs for ever

[arth1]

I'd love to see a mainstream Linux distro adopt Sun's ZFS as its default filesystem.

Oooh, imagine a Zettabyte pr0n collection!

Re:xfs for ever

[arth1]

There have been too many reports in the last couple of months of people whose machines have lost power, and booted up, only to find that every file on their XFS filesystems has been filled with zeroes.
That's what backups are for. Seriously, with XFS you run a very real risk of zeroing out a file if the file system isn't shut down properly. But with reiserfs, you run a very real risk of losing the file system. In over a decade, I've never seen that happen with XFS -- only zeroed out files.

Regards,
--
*Art

Re:xfs for ever

[cortana]

Actually, the zero-filled files is a misfeature of XFS. Having a UPS will not save you. There are two XFS problems:

  1. Power loss can destroy your filesystem. Solution: do not use XFS or ReiserFS without a UPS.
  2. An unclean shutdown can leave you with zero-filled files. AFAIK this is a design flaw in XFS or, depending how you look at it, a tradeoff of data integrity for performance. If you don't like the tradeoff then your only choice is to use another filesystem.

Source: http://linuxmafia.com/faq/Filesystems/reiserfs.htm l

Re:xfs for ever

[eneville]

There have been too many reports in the last couple of months of people whose machines have lost power, and booted up, only to find that every file on their XFS filesystems has been filled with zeroes.

i cannot see why the filesystem would be zeroed, perhaps a while, which would make sense, since a open(), seek(), write(), close() can result in a file full of zeros, if only part of it is written then that makes perfect sense.

are there any documents or reports of the file system being zeroed? i've never had a problem with it in the years i've been using it as my primary file system.

Re:xfs for ever

[eneville]

what about a shared pr0n collection with afs?

Re:xfs for ever

[udderly]

That's what backups are for. Seriously, with XFS you run a very real risk of zeroing out a file if the file system isn't shut down properly.

OMG, are you kidding? If it was NTFS or FAT, people on /. would be going crazy about it. It would be more famous than the BSoD.

Re:xfs for ever

[Anonymous Coward]

I have used ReiserFS for a long time, on many different systems and different circumstances.
Systems have lost power, and maybe a few times systems have crashed. Systems had uptimes over a year, have been working for several years until they were rsynced over to new hardware and continued working.

In all this time I have never seen anything go bad on a ReiserFS.
There is only one thing I did not like: one time I had a bad block (track, really) on a disk, and this of course resulted in some data loss. But worse, ReiserFS refused to work around the damage and the manual page declared that I had to buy a new disk. I found this sub-optimal but I did not lose other data than what was in the bad area.

Are you sure there is a problem? I have not seen it. Could it be bad hardware? Maybe the victims had read/write errors more often? (I normally use RAID-1 for all filesystems except the one mentioned above)

Re:xfs for ever

[cortana]

In the defence of the ReiserFS, if a disk reports a bad block to the operating system then it means that its internal supply of spare blocks (that are used to transparently replace bad blocks) has been exhausted, and that the disk should be replaced immediatly.

For example:

# smartctl -A /dev/hdg
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  5 Reallocated_Sector_Ct   0x0033   253   253   063    Pre-fail  Always       -       0

This disk hasn't yet had any bad blocks. As the disk ages and blocks go bad, RAW_VALUE will go up, and VALUE will go down. When VALUE <= THRESHOLD then there are no more spare blocks and bad block errors will be reported to the operating system. At this point the disk should be replaced.

Re:xfs for ever

[Chatz]

Can you please point me to a single instance of a complete filesystem being filled with zeros?

Files that have been grown and written into cache may contain zeros because the linux vm did not ask XFS to write that data out yet. The application (and user) always has the option of sync'ing the file to push it to disk straight away.

Re:xfs for ever

[RAMMS EIN]

``Are you sure there is a problem? I have not seen it.''

The problem is, the fact that you are not experiencing any problems doesn't mean they aren't lurking in a corner somewhere. Once you find your filesystem has eaten your precious work, you know there's something wrong, but the fact that it hasn't done so (yet) doesn't prove anything.

Now, if you search the Tubes, you will find that some people have experienced problems with ReiserFS. Clearly, it hasn't always been rock solid. Are these problems fixed now? Let's hope so. Does that mean ReiserFS is bug free now? Probably not.

Of course, the same is true of pretty much any other filesystem, as well. In my personal experience, ReiserFS has been the most reliable _and_ the fastest filesystem I've used, but that's just one person's experience.

Re:xfs for ever

[Reverend528]

only to find that every file on their XFS filesystems has been filled with zeroes.

A similar thing happened to me, but it was just one file, /dev/zero. I even tried switching filesystems, but that didn't fix it.

Re:xfs for ever

[Rob Kaper]

What good is a UPS going to do in the case the machine powers off because of a problem with the power unit, a motherboard short circuit, and so on? Any filesystem with serious data loss on a power failure is not acceptable, period.

Re:xfs for ever

[Anonymous Coward]

That's exactly right. Reiser-bots (or any filesystem zealots) should remember this: EXT2/3 are virtually bullet-proof. They are used in a HUGE numbner of systems and have been tested over and over and over again in a vast array of hardware and conditions.

EXT3 works. Some filesystems do niche tasks better (very large systems, for example, might be better with GFS)... but EXT3 is a workhorse. It does well in all conditions and for general purpose use it is unbeatable. If you don't know what you are doing, do not listen to zealots. Stick with the distro default -- which is EXT3/

Re:xfs for ever

[Anonymous Coward]

In the defence of the ReiserFS, if a disk reports a bad block to the operating system then it means that its internal supply of spare blocks (that are used to transparently replace bad blocks) has been exhausted, and that the disk should be replaced immediatly.

For writing this is mostly true, but not really for reading. When reading a sector the disk will always report it as bad if it is bad, regardless of how many spare sectors there are left.

There are two cases when a sector will be reallocated:
1. When you write to a known bad sector.
2. When you successfully read a sector that is about to go bad.

Re:xfs for ever

[chill]

That would be the A to Z collection of pr0n!

Re:xfs for ever

[Anonymous Coward]

I've been using XFS on IRIX boxen for more than 10 years and never in that time have I experienced data loss. Must be a Linux version only problem. The JFS port is a half baked version lacking the robustness of the AIX version. Maybe it's the same with XFS.

For junk hardware like x86 I usually stick to BSD and UFS2.

Re:xfs for ever

[dobestpossible]

So true, so true. A double standard is shown by Linux fans to "welcome the open source overlords!" whatever that means.
  I am just now learning about the Linux OSs and haven't tried the ReiserFS or XFS yet, so can't really add anything to the post *sorry*

Re:xfs for ever

[rg3]

It's interesting that you mention that. Some time ago, I used ReiserFS as the filesystem on my laptop computer (I only have one partition, not counting swap). The performance was alright and it always took some seconds to mount the partition (this is a known thing for ReiserFS). So, more or less, my experience had been fine. One day, I was trying to view one JPG file and the program was unable to open it, so I wondered why. After examining the file, I found out that while the file size was alright, its contents were all binary zeros. I discovered similar things for a handful of files in my system, many of them in my home directory, I supposed because that's where the biggest part of the disk data is located and if a problem arose, it's probably going to be there.

At the beginning I suspected something had gone wrong while copying the data to an external USB hard drive and back to the newly formatted ReiserFS partition. But, some weeks later, I discovered a similar situation in a file I had created recently (after the data move), and that had been available there for many days. I am only a desktop user and I lack evidence on what caused this, but I tested my harddrive to see if it had bad sectors or behaved poorly for some reason, and nothing turned up. I fsck'ed the partition and everything was alright. I suspected this problem was due to ReiserFS, so I took the decision of switching back to ext3 with dir_index activated, and the problem hasn't reappeared again. I suspect I hit a bug in the ReiserFS code, and I lost my data in one or several of those ocasions when I left my laptop alone for some time and it powered off suddenly when it ran out of battery. This happened more times since the switch to ext3, but I haven't lost any more files since then.

I know this can be a particular case which may not represent the behaviour of ReiserFS, but as I read your comment I thought I had to share my experience too.

Re:xfs for ever

[fire-eyes]

Yup, no surprise there. XFS caches writes very agressively in ram, around 50MB if not more, for long periods of times. So it "feels" fast but really isn't in some aspects.

So you pop the power off and *wham* bye bye cached data. This is definately not any kind of fun.

XFS was written for environments where the power just dooes not go out -- datacenters, people with a very good UPS etc. I generally recommend XFS for people with lots of large files, but if they don't have a good power backup, I change my mind.

Re:xfs for ever

[diegocgteleline.es]

AFAIK this is a design flaw in XFS

No, this happens because it's the way XFS does journalling.

XFS journalling isn't as good as the one in ext3, from users' POV. Ext3 default journaling mode takes care of the relationship between metadata and the data associated to that metadata (and here let me remember that journalling/softupdates is a way to avoid corruption of the *metadata*, if you lose data because of a power cut that's fine, but it's not fine that the filesystem gets damaged and needs fsck because the metadata got corrupted)

IOW: when ext3 is going to write metadata to the disk, it looks first to the dirty data cached in the memory and writtes the data *before* it writes the metadata.

XFS journaling, in the other hand, does *not* care about writing the data before the metadata. Why? Well, because journalling is about keeping the metadata safe so you don't need fsck. This means that in case of a power cut, XFS may leave the contents of a "file" (metadata) unscycrhonized with its data. Because of that, the metadata may be pointing to random free zone of the disc with confidential information (passwords) which was deleted but it has not been overwritten, so XFS sets it to zero for safety. Ext3, on the other hand, will never left your data "unscychornised" with your metadata. The file may get corrupted because the program that was manipulating it was stopped in the power cut, but the relationship between the data and the metadata is always coherent.

Ext3 journaling mode may be considered an "extra", and it *does* pay a performance disadvantage because of this. If you want ext3 to behave like xfs (and get better performance), mount your fs with the mount option "data=writeback". Reiserfs in the other hand historically had a similar journaling method as XFS (just like JFS), but the suse guys created a journaling mode similar to the default one in ext3 which AFAIK is not enabled by default (at least on mainline) and gets enabled with "data=ordered"

Is the XFS journaling mode worse? Well, for desktop users, who would rather have syncronized their data and their metadata, clearly yes. This is why XFS is just not the best FS for desktops - its a wonderful FS, but just not "optimized" for desktops. NTFS journaling does the same that ext3 does, BTW, and it's for a reason.

Re:xfs for ever

[Anonymous Coward]

No no, this is not true! The disk can have read errors due to bad blocks, that disappear when the blocks are written to and then re-allocated.
In fact this was true for that particular disk. After writing to the bad spot the problem went away. But ReiserFS and -fsck did not want to try that.

Re:xfs for ever

[AusIV]

Neither xfs nor jfs partitions can be reduced. This may not be a big deal to compaines who just add disks and expand their partitions, but I know that I lost about two hundred gig worth of data that would probably still be around if I could have reduced my jfs partition. After that I tried to install ReiserFS then Reiser4, and after a little bit of trouble with those, decided I'd use Ext3 because it just works. Even if its performance isn't as great as some other file systems, I don't know too many people who have lost data because of flaws (or "features") of the filesystem.

Re:xfs for ever

[byolinux]

Sounds like bad hardware to me. Send your hard disk to me for analysis.

Re:xfs for ever

[spyfrog]

The only two times I ever have lost data due to data corruption in Linux is on system and disks that used EXT3. Since then I have used ReiserFS and never had any problems.
My trust in ext3 is so low that it can't be reestablished. That two of my friends have had the same problem with ext3 and never with reiserfs doesn't help me getting more confidence in ext3.

Re:xfs for ever

[RevDobbs]

OMG, he's not kidding.

And... so what? It is a "high performance" file system, not a safe file system. Use it for application cacheing, tmp space, or something else where file integrity after a reboot does not matter. It is just a tool, another option when using your unix-like computer.

Re:xfs for ever

[Skater]

I had exactly the opposite experience: ReiserFS did something VERY scary for me, while ext2/3 have always been fine.

When I was using ReiserFS on my laptop I had a file get, for lack of a better term, "stuck". I couldn't edit it, couldn't delete it, basically, I couldn't touch it, even as root. It was not marked immutable, or anything like that - I tried everything I could think of, I asked around to other Linux users, etc. Eventually, I had to format the drive to "unstick" that file. To me, that's a pretty scary development in a file system.

Re:xfs for ever

[schon]

Reiser-bots (or any filesystem zealots) should remember this:
[snip religious propounding of Ext2]

Do you know the definition of hypocrisy?

Re:xfs for ever

[Bert64]

I have been using XFS for years on both linux and IRIX...
It has proven itself very stable, and has caused me no problems (the only problems i've ever had were due to physical hardware failure)..
The biggest issue i've had with EXT2 is that when it does an fsck at boot, if theres a problem it sits there waiting for input, which is useless on a server with no console attached... EXT3 still does this too, it still demands to be fsck'd every few months.

Re:xfs for ever

[MightyYar]

Any filesystem with serious data loss on a power failure is not acceptable, period.

That may be true in your application, but some might be willing to take that risk in exchange for performance.

Re:xfs for ever

[jazman_777]

Slashdot: News for Nerds. Stuff that Matters. Anecdote Battles.

Re:xfs for ever

[UncleTogie]

{(the only problems i've ever had were due to physical hardware failure)...}

THAT is the reason I'd avoid both XFS and ReiserFS. Hardware fails. This is a given; c'mon, hard drives have an MTBF rating for a reason. I'll go with a FS that can handle real-world scenarios for people withOUT huge datacenters.

Re:xfs for ever

[BeeBeard]

I've had still another experience. Ext3 and ReiserFS have been solid as a rock under Linux for years now, and I've only run into trouble when I get brave and try to read and write to ext2 and ext3 partitions from Windows. Something about how windows (not surprisingly) does not respect the file permissions.

Re:xfs for ever

[ArbitraryConstant]

"Any filesystem with serious data loss on a power failure is not acceptable, period."

I seriously doubt that I would care if my squid proxy box lost the filesystem with the cache on it.

It is entirely application-dependant.

Re:xfs for ever

[WilliamSChips]

I've never had any problems with either ext2 or ReiserFS that were filesystem problems rather than hardware problems or really weird kernel bugs.

Re:xfs for ever

[arivanov]

Quite correct.

I have quite a few TB+ volumes on XFS and I have not experienced this problem yet. Granted, I have not had a single case of power failure shutdown on any of these servers. They are all on UPS and the average time between reboots is 200+ times the backup interval (they are backed up nightly and the average uptime is 200+ days).

Nice to know that this can happen anyway (I have yet to see it in nearly 4 years of using XFS in production).

Re:xfs for ever

[Anonymous Coward]

J was murdered by being nailed to an X. These days in the U we murder murderers (and alleged murderers) by strapping them supine to an X and injecting them with chemicals. But we say we follow the teachings of J.

Re:xfs for ever

[perlchild]

You can remove this behaviour with the tune2fs utility, while the partition is unmounted. Most linux distros do this by default on ext3. On ext2, for obvious reasons, it's not recommended.

Re:xfs for ever

[Skater]

You expect us to argue with facts? That's crazy! :-)

That reminds me: We haven't had a good hard disc brand anecdote war here lately... It's about time to stir another one up! :-)

Re:xfs for ever

[Znork]

"ZFS as its default filesystem."

While ZFS looks impressive on a featurelist, I really dont like the monolithic one-size-fits-all cram approach. The current linux capabilities with the device-mapper and stackable block devices are vastly more flexible in the long term.

Re:xfs for ever

[Anonymous Coward]

XFS has trashed at least 3 filesystems at my site (and they were fairly recent versions of the kernel, too). It seems to be very incapable of dealing with the number of power outages we have here.

JFS & ext3, on the other hand, have been rock solid. (In fact, ext3 was shown, in a test last year, to be the most stable filesystem in the event of power outages.) As a result, we use ext3 for the /boot partition, and JFS on our /home, /opt, /usr directories. Works like a charm; no problems in over 4 years.

Re:xfs for ever

[Ash-Fox]

Now, if you search the Tubes, you will find that some people have experienced problems with ReiserFS.

Lost data

Re:xfs for ever

[Anonymous Coward]

Brilliant!

Re:xfs for ever

[und0]

According to an old email from Ts'o, not only they suggest to run XFS with an UPS but SGI hardware was modified to mitigate damages in case of black-out using big capacitors and, at kernel level, was added a power-fail interrupt to Irix. http://linuxmafia.com/faq/Filesystems/reiserfs.htm l

Re:xfs for ever

[kimvette]

It seems to me that is more of an issue with the ext driver for Windows, and not Windows itself; it is doing what the filesystem driver does not forbid. IMHO the fault does not lie with Microsoft there, as much as I'd LOVE to blame them. :)

Re:xfs for ever

[tcgroat]

What good is a UPS going to do in the case the machine powers off because of a problem with the power unit, a motherboard short circuit, and so on?

The UPS covers the power line problems, which are the leading cause of system outages. To protect against the less frequent hardware failures you need properly engineered redundancy for every critical component. That's why "enterprise-class" data storage costs so much more per GB than the disk drives on sale at retail store. An alternative is to not use any write-behind caching. The performance loss varies, depending on the write/read ratio. The average desktop system doesn't gain much performance from write caching, because the most time consuming disk activity is loading multi-megabyte executables and existing data files, not writing and updating files. That also means they're at less risk in any case, because there is less unwritten data in the cache to cause problems. An online transaction system, on the other hand, does continuous file updates and little application loading. These need professional quality hardware (and software!) to deliver the expected level of reliability.

Re:xfs for ever

[finiteSet]

One time a rm -r in the wrong terminal erased nearly a month's worth of work. Obviously, entirely my fault (one several levels). Because I almost universally use ReiserFS, I was tremendously releived to find several stories of successful file recoveries (such as this one). Unfortunately, I soon realized I had used ext3 for this particular partition...

Q: How can I recover (undelete) deleted files from my ext3 partition?

Actually, you can't! This is what one of the developers, Andreas Dilger, said about it: In order to ensure that ext3 can safely resume an unlink after a crash, it actually zeros out the block pointers in the inode, whereas ext2 just marks these blocks as unused in the block bitmaps and marks the inode as "deleted" and leaves the block pointers alone. Your only hope is to "grep" for parts of your files that have been deleted and hope for the best.

Though I have a much better backup system now, I still avoid ext3 at all costs. As careful as I try to be, I know I'll slip up again sometime.

Re:xfs for ever

[alba7]

If you don't know what you are doing, do not listen to zealots. Stick with the distro default -- which is EXT3/

If you don't know what you are doing, you should stick with the default "distro".
And Microsoft recommends NTFS.

Re:xfs for ever

[Anonymous Coward]

Well I know. And to any normal person, that wasn't "religous expounding" (that's why you snipped it... rather spoils your posts).

It was simple commonsense... and you're a fucking idiot. You did read the rationale for SUSE switch, didn't you? Or do you think he's a zealot too?

Pff, like I said. File system zealots are a fucking menace. You probably cause more problems for newbies than anyone else, apart from distro zealots.

Re:xfs for ever

[Anonymous Coward]

Interesting... it has never happened to me even though I have moved many terabytes of data through ReiserFS systems.
Maybe a laptop is different because it could go to suspend mode at some inconvenient moment.

Re:xfs for ever

[BillyBlaze]

Try adding nodev to the options column for that filesystem in /etc/fstab.

Re:xfs for ever

[Anonymous Coward]

JFS is a few years older than XFS, I'd have to say it's probably a bit more mature than XFS.

Re:xfs for ever

[From A Far Away Land]

I find nothing pounds "performance" into the ground better than a loss of power and then data. If you're playing games, then data loss is acceptable, but if you need what the computer is keeping, then data loss is End of Game.

Re:xfs for ever

[NormalVisual]

The U just doesn't remember the teachings of J because K pulled out the flashy thing and used it on them.

Re:xfs for ever

[DrXym]

What should be considered is that anyone absolutely seeking out a high performance filesystem is capable of installing it for themselves. For everyone else, what is important are reliability, self-repair, ease of maintenance, code maturity, general desktop performance, availability of tools and good information for when things go wrong. Nothing touches ext3 for those things. I have no idea what would be best for servers, but I suspect even many of those could function quite happily with ext3. For those that don't, Novell should build other filesystems into its kernel but leave it up to admins to decide what to do.

Re:xfs for ever

[instagib]

That might be funny. But it is the same with all Linux FS discussions: for every major FS, there are people who never had problems, and others who "lost all their data". Which means: one cannot choose a FS based on real world experience. So what do you base your decision on? In the end, it's reiserfs3 for lots of small files, xfs for large (100MB+ ?) files, and ext3 for the rest. Or isn't it?

Re:xfs for ever

[udderly]

So, let me get this straight, a "high-performance" file system is one that screws up your files. And somebody actually modded that "insightful."

From wikipedia: "Fanboys" remain loyal to their particular obsession, disregarding any factors that differ from their point of view. They are also typically hateful the opposing brand of their obession regardless of its merits or achievements.

Sound like the guy in the mirror?

Re:xfs for ever

[badfish99]

I work with machines that get abused like this a lot: naive users think that the way to fix a problem is to power cycle the machine. And I've seen a lot or ext3 file systems that are utterly screwed up beyond repair as a result. I've never had a similar problem with reiserfs.

Re:xfs for ever

[cptnapalm]

A lot of comments mention that they find files which are zeroed out. To this I would like to add my own odd experience with xfs. I had formatted an external drive with xfs; it hosts all my video files.

X (not the FS) began to lock up and I had to do a hard reboot a few times. This is not good for filesystems.

A lot of my video files were marked as zeroed out. But the odd thing is that if I moved them from the commandline to another location, everything was good again.

I don't know enough about filesystems to say why that was the case, but if anyone finds that they can't use a file under xfs, try manually moving it and check it. What can it hurt?

Re:xfs for ever

[Wdomburg]

I've had terabytes of data stored on hundreds of ext3 filesystems for years. The only data loss we've had has been due to hardware failure or user error.

Re:xfs for ever

[Jerry]

Add me to your list of people you know who have had BAD experiences with EXT3 and have found ReiserFS V3 totally reliable since SUSE included it in 6.4.

Novell can expect a RASH of complaints about slow response and data loss with Ext3.

Re:xfs for ever

[quinnharris]

In theory and from my experience with reiser4 for nearly 3 years, it does not have this problem. But it is extremely lazy. By default reiser4 can take up to 10 minutes to sync data to disk. So you will get a perfectly consistent file system after a crash but it might be a bit out of date.

Also note the reiser4 is extreally dependent on data being written in a specific order (like softupdates). I had some fs corruption issues on my laptop with reiser4 until I disabled the write cache. I believe that hd wasn't respecting write barriers or whatever it is (if anything) for that drive. I have never had a problem like this with my SCSI Cheetahs.

I have experienced this individual file corruption with reiser3.

Re:xfs for ever

[MightyYar]

Yeah, but maybe you have a big render farm set up and you need speed and large file size support. If one node goes down and loses all of its data, that's no big deal - just let the render farm re-render. Scientific computing is similar, as is data mining. There are plenty of applications that might favor some performance over reliability.

Re:xfs for ever

[udderly]

I know that this is going to sound ass-kissy, but that has to be one of the best posts I have seen on /.

Very informative with none of the "you're a dumbass because you don't know this" kind of crap typical on /.. Also, well organized, very readable and generally grammatical. Bravo!

Re:xfs for ever

[Bert64]

In the cases of physical hardware failures, i`ve lost data with ext2, ext3 and xfs...
Sometimes the disk locks up, and you can let it cool down, then bring it back up and you might get enough time out of it to copy the data off, when this has happened i've never had worse luck with ext2/3 than xfs...
And sometimes the disk dies completely, and will never power up again, when this happens it doesn't matter what filesystem you have, your data is GONE short of specialist recovery hardware.

Re:xfs for ever

[vtcodger]

***OMG, are you kidding? If it was NTFS or FAT, people on /. would be going crazy about it. It would be more famous than the BSoD.***

You have it nailed, bro. FAT -- by design -- rarely hoses files although it may tie up blocks of storage until they are released by CHKDSK/SCANDISK. Ext2 on the other hand will lose inodes if you cough while in the same room with the server. Try suggesting around here that FAT with umsdos just might be a better FS for a home Linux user with an older computer than ext2 and you'll be treated like you have three heads -- all ugly --and have just arrived from the Gamma Quadrant.

Fortunately, that's no longer an issue since ext3 is journalled and doesn't lose files. (NTFS is journalled also, but if you think I'm going to keep data I care about in a proprietary Microsoft FS ...)

Re:xfs for ever

[fbjon]

Plenty yes, but none of those pertain to the majority of users.

Re:xfs for ever

[theLOUDroom]

What good is a UPS going to do in the case the machine powers off because of a problem with the power unit, a motherboard short circuit, and so on? Any filesystem with serious data loss on a power failure is not acceptable, period.

How about a diskless workstation, that isn't going to have ANY data left after power loss, regardless of file system?

It's really foolish to make blanket statements like that.
Perhaps it's not useful to you, but consider the situation where you've got hundreds of redundant servers rebuilding their dataset every 24 hours. Even a 1% speed increase is going to more than justify the cost of keeping one extra server around in the case of filesystem corruption.

What you're doing is like looking at a Ferrari and saying "That's stupid and useless. It's much less reliable than my Toyota."

Although I would never run my personal workstation on this filesystem, I can see obvious cases where it could prove useful.

Re:xfs for ever

[obeythefist]

Often a true word is spoken in jest, there seems no choice in the windows world.

Having never seriously contemplated which filesystem I want to use, and living mostly in a Windows world where your choice between FAT32 and NTFS is simple, and WinFS is not to be seen, I beg to ask some questions.

How do these file systems compare to NTFS? If NTFS was open source, would people use that instead of xfs/etc?
Are there windows drivers for these filesystems so we can mount them in Windows? Should these be used instead of NTFS? What's the lowdown?

Re:xfs for ever

[sirsnork]

Short version is, unsurprisingly, it's good at some things and very very bad at others http://tortoisesvn.net/node/41

Re:xfs for ever

[jZnat]

NTFS gets fragmented for one, so there's a big problem. I don't know if it's journaled either.

Re:xfs for ever

[arth1]

JFS is a few years older than XFS, I'd have to say it's probably a bit more mature than XFS.

Not so:
XFS: 1994 (IRIX 5.3 XFS)
JSF: 1999 (OS/2 Warp)
ReiserFS: 2001 (Linux 2.4.14)

Re:xfs for ever

[slittle]

This happened to me after the 2.6.18 upgrade, on two machines. First time an abnormal shutdown (of which there have been plenty) has required xfs_repair at all.

I also had to compile xfsprogs from source because Debian Sta(b)le is too fucking old to fix it, but it was mostly alright.

Re:xfs for ever

[Enahs]

Yeah, I think people actually think that JFS for Linux is the AIX JFS, when really it's a close cousin to the JFS that was developed for OS/2.

I keep meaning to try it out; it sounds great, but people get so wound up over the EXT3 vs. ReiserFS that other systems tend to get forgotten. Plus people went all giddy when XFS was released for Linux...JFS just wasn't as hyped-up as the others.

Re:xfs for ever

[shadowmas]

"I get brave and try to read and write to ext2 and ext3 partitions from Windows. Something about how windows (not surprisingly) does not respect the file permissions."

This is not exactly a problem with windows. it's a problem with file system driver not checking for permissions. perhaps this is not possible since windows and linux permission structures arent exactly the same. The driver wou;d also need to find the /etc/users file to decode the linux users list (which again might not mach with windows users)

the same problem exists in linux ntfs drivers which also doesnt respect windows permssions, i think for the same reasons.

and just in case you didn't knew already there are also a windows driver for the reiserfs (don't have the url with me right now)

Re:xfs for ever

[Anonymous Coward]

ntfs is journalled. it can get fragmented, but so can ext2/3, xfs, jfs, reiserfs, etc.

fragmentation is not a big problem for any of these since they all are fairly effective at remediating fragmentaiton. (xfs is perhaps the best at it, fwiw)

Re:xfs for ever

[SanityInAnarchy]

For the record, a UPS is something every computer should have in the first place. Whether it's two hours ago or five seconds ago, power fail without a sync means you WILL lose data.

Also, I have never seen XFS or ReiserFS need a FSCK. I've seen Reiser4 need one, but only on a box that I routinely overclock.

Re:xfs for ever

[LittleBigLui]

Well, with windows systems you don't have much choice in your file systems, do you?

Certainly, if you have a bunch of file systems to choose from (ReiserFS, ext3, XFS, JFS, etc.), one of them having this bug/feature is certainly not as troublesome as if you only had two file systems to choose from (FAT, NTFS) and those two were already different enough that you didn't have much choice anyway. (FAT is old and lacks features, but is very universal. Hence, great for USB sticks, MP3 players, partitions used for swapping data between windows and other systems on multi-boot systems, etc. NTFS for everything else.)

Hence, no double standard at all.

If, say, NTFS had this characteristic, millions of users had the choice of risking losing their data on power loss or using FAT only.
Since XFS has this characteristic, users that don't want to take that risk (dpendending on the application, it might be worthwile to actually take it) still have plenty file systems to choose from.

Re:xfs for ever

[RevDobbs]

XFS makes very specific trade-offs in integrity for the sake of speed.

Some may say that mySQL does the samething, versus an ACID-compliant database like postgres.

But like I posted, it is just a tool, there is nothing to get excited about. If it is not the right tool for your job, so be it.

Re:xfs for ever

[RevDobbs]

And no, troll, a high performance file system is not one that screws up your files.

I gave specific examples of cases where file integrity across reboots does not matter. You, on the other hand, chose hyperbole and personal attacks as your argument.

Re:xfs for ever

[Ginger Unicorn]

That isn't getting it straight at all. Thats just a bunch of handwaving. His point was that XFS omits some of the journaling features that ext3 includes so that it can perform faster. And then he gave perfectly acceptable examples of where this might make XFS the better choice.

From wikipedia "A straw man argument is a logical fallacy based on misrepresentation of an opponent's position. To "set up a straw man" or "set up a straw-man argument" is to create a position that is easy to refute, then attribute that position to the opponent."

basicly he was being perfectly reasonable, and you are a sarcastic stroppy little prick who is making shit up to try and save face.

Re:xfs for ever

[cortana]

I use 'ext2fsd' to mount ext2/3 filesystems in Windows. Read-only for ext3, unfortunately.

Re:xfs for ever

[udderly]

I gave specific examples of cases where file integrity across reboots does not matter. You, on the other hand, chose hyperbole and personal attacks as your argument.

Okay, so I did. Sorry, that was childish. I guess that it was in response to what I perceived as a mocking tone in your first reply (you know, the "OMG, he's not kidding" thing).

In any event, my original point wasn't a criticism of XFS. I was merely commenting that the general /. audience would be far less accepting of XFS's limitations/optimizations/features if it were a Windows FS. That is all.

Re:xfs for ever

[udderly]

basicly he was being perfectly reasonable
While I will admit that he was being reasonable while doing so, his OP was a red herring (since we're getting into logical fallacies). It did not address the point in my OP at all. See http://en.wikipedia.org/wiki/Ignoratio_elenchi#Red _herrings. Not that it justified my little outburst, mind you.

and you are a sarcastic stroppy little prick who is making shit up to try and save face.
Generally guilty as charged (see breakdown below), although you didn't have to sugar-coat it like you did.

1=Not me at all-------2--------3---------4---------5=Me exactly
sarcastic=5 (I'm working on that)
stroppy=4 (I had to look that one up)
little=2 (actually, I'm pretty big)
prick=4 (see sarcastic)

Re:xfs for ever

[udderly]

Good point(s). I basically concede the issue. My only caveat is that I still think that there would be more criticism--all things being equal--if it were a Windows OS. Even if Windows could use all of the other OSes as well.

Re:xfs for ever

[LittleBigLui]

Probably. But that's because windows sucks. :)

Re:xfs for ever

[R2.0]

"three heads -- all ugly --and have just arrived from the Gamma Quadrant"

Now, if you have 3 tits -- all big -- and are from Mars, you're set.

Re:xfs for ever

[Laur]

How about a diskless workstation, that isn't going to have ANY data left after power loss, regardless of file system?

Uh, if it's a diskless workstation, what exactly are you formatting with XFS or Reiser? How is a diskless workstation at all relevant in a discussion about the merits of various filesystems?

Re:xfs for ever

[TommydCat]

I think it may be a bug, but with XFS I noticed all these "." and ".." files all over the place. After I tried to removed them, my whole filesystem was zero'd out!

Methinks this bears investigation...

Re:xfs for ever

[theLOUDroom]

Uh, if it's a diskless workstation, what exactly are you formatting with XFS or Reiser?

RAM

How is a diskless workstation at all relevant in a discussion about the merits of various filesystems?

RAMdisks

Rats first and Captain last

[Gopal.V]

At least that's what happens to a sinking ship. A maintainer going missing does not quite instill the users with confidence, especially when it is happening due to reasons other than flagging interest. Most commercial distributions have SLAs which sort of work against such brilliant work by an individual contributor - they just can't depend on the whims of a person or his fate.

One of my friends once told me that "Extraordinary hackers are people with socially acceptable problems". In fact to achieve what they feel they must, a lot of them give up a lot - health, social lives and financial security. But because a few do that, does not mean FOSS programmers are crackpots. And I say this as a son who's home (which I can because my commits go to a public CVS) watching over a sick father.

So as understandable as it is that commercial vendors might want to switch away, but that doesn't mean anyone gets to shine a torch or make jokes into somebody else's darkness.

Re:Rats first and Captain last

[Anonymous Coward]

"What you learn, when you read works like Novum Organum by Sir Francis Bacon, is that science is about being a blind man with a stick, and he who most persistently pokes blindly ahead of him, contributes the most to our understanding of the Universe, though only if he is willing to accept what the poking tells him that he does not want to be true. I am not as qualified or clever as our competition, and we aren't as well funded, but we are much more persistent and rigorous. That is not what I wanted to believe would be my contribution to the field when I was a boy, but so it is."

-Hans Reiser

Re:Rats first and Captain last

[zdzichu]

Almost right, but the fact is Hans Reiser wasn't reisersfs3 maintainer. He long ago declared version 3 was dead and only reiser4fs worth using. reiserfs3 was maintained mainly by one guy in SUSE, who became fed up with it. And rightly recomended going to ext3/ext4.

Just BTW, I am using reiserfs3 on my system and I thinking about migration to some FS with future.

Re:Rats first and Captain last

[Anonymous Coward]

At least that's what happens to a sinking ship.

I don't think it is... or at least, SUSE's decision isn't motivated by the recent unpleasantness over Reiser's arrest. Back when ext3 was under development, I remember being swamped by the usual Reiser-bots claiming that it was the future... that ext3 was just crap built on crap etc. I kept trying to tell them that ext3 was essential, that all those ext2 systems had to be supported and that the up/downgrade feature of ext3 was *killer*. People don't want to wipe and reformat. I urged them to only use Reiser for stuff like newsfeeds or temp filestores (since it worked well for lots of small files)... and that besides the manifest incompatiblity with ext systems, it had shitty recovery tools as was brittle in the face of hardware problems. None of them listened and they carried on screeching like harpies whenever anyone mentioned ext file systems.

And now, finally, SUSE comes to their senses. NOTE: this isn't a criticism of Reiser -- it has its advantages. Just don't base any critical long-term systems on it. Your default choice should be ext3. And don't listen to the legions of zealots who say otherwise. Most of my advice from 5 years ago still applies today.

Re:Rats first and Captain last

[ObsessiveMathsFreak]

And I say this as a son who's home (which I can because my commits go to a public CVS) watching over a sick father.

I think your father would recover a lot faster if you weren't by his sickbed clacking over your keyboard and humming PC. Word to the wise, if you're in deep hack and he asks for water, don't pass him the Mountain Dew.

Re:Rats first and Captain last

[ScrewMaster]

I think your father would recover a lot faster if you weren't by his sickbed clacking over your keyboard and humming PC.

I know you're just being funny, and I don't really mean to pick on you, but that's rather unkind. As someone who spent three years doing the same thing full-time before my father died, I'd say his Dad should be bloody well thankful that he has a son willing to take care of him, clacking keyboard or otherwise. Heck, when I moved back in to my father's old house I moved my entire multinode Wildcat! BBS back in with me. It was sixteen processors, plus the file server and a couple of miscellaneous workstations. He thought it was cool (all those CGA monitors with text scrolling by looked like Mission Control.) I remember that system generating so much heat that we turned the furnace way down in winter.

Taking care of a sick person is difficult at best, even if you're trained for it, and if that individual happens to be a parent it is doubly hard, believe me. What Gopal.V is doing is very uncommon in this day and age (at least in the U.S.) and is to be commended. Parents are too often treated as disposable commodities that end up dying by degrees in a nursing home surrounded by strangers, because family members who should know better are too "busy" with their "responsibilities", or just too lazy and self-important to be bothered caring for the previous generation.

Sorry for going off-topic.

Re:Rats first and Captain last

[Antique Geekmeister]

Don't bother. "Journaling" is fine when it protects your file system, but when it actively imperils your filesystem due to mis-handling of hardware failures, it's not worth using. The big advantage ReiserFS had for a long time was being able to handle 50,000 files in a single directory, such as might occur in an active news spool or very clogged Maildir mail directory. But ext3 has been able to deal with that since the introduction f "htrees" in the 2.6 kernel: there's simply no reason to risk your filesystem with ReiserFS anymore, unless the very modest performance benefits is wrth risking your data.

ReiserFS, like db4 for databases, has lost its original core developers and has become increasingly unstable in its search for new features.

Re:Rats first and Captain last

[glitch23]

Actually Hans didn't go missing. It was his wife (ex-wife? I forget what I read). But he was arrested upon suspicion of her disappearance/murder which could be worse compared to if he was just missing.

Explain to me, please

[Anonymous Coward]

it claims ext3 is more stable and will 'soon' match performance with the newer ReiserFS 4

ext3 will match reiserfs4? how? when? are they talking about ext4?

Re:Explain to me, please

[Anonymous Coward]

yes, i was wondering why they were talking about matching performance to reiserfs 4. they probably mean ext4.

conversion

[Anonymous Coward]

Hopefully someone will write a conversion utility...
Right now, it is very unpractical to change existing systems from ReiserFS to Ext3. It is either "install from scratch" or "dump and restore with enough trickery to get the correct fstab and initrd on the next boot".

Re:conversion

[sydneyfong]

Can you even dump in one fs format and restore in another?

Re:conversion

[Jessta]

why would you convert your file system? There is no reason to unless you are actually having issues with it.

Re:conversion

[Ant P.]

Yes, in fact someone made a command to do just that: `cp`

Re:conversion

[thrillseeker]

why would you convert your file system? There is no reason to unless you are actually having issues with it.

Why would anyone living on the west coast of the US quit eating fresh spinich - just because he's read of others getting ill from it doesn't mean he will.

Re:conversion

[Anonymous Coward]

tar? cp -R?

Old news

[lintux]

This news comes shortly after Hans Reiser's arrest

That news was this week. This news from SuSE, however, is very old already and apparently they indeed decided about this before Reiser got arrested.

It's also interesting how people now explain the blood on Reiser's shirt in this comic, while this comic also predates this whole arrest story. :-)

Re:Old news

[Monoman]

Mod the parent up. This was pretty well known before the Reiser arrest.

Re:Old news

[Anonymous Coward]

This 'story' should be pulled. It is not true to say that they announced this after the arrest - therefore there is no story.

Re:Old news

[Megane]

It's also interesting how people now explain the blood on Reiser's shirt in this comic, while this comic also predates this whole arrest story. :-)

Sheesh. Too Soon.

Re:Old news

[ralphdaugherty]

This news from SuSE, however, is very old already and apparently they indeed decided about this before Reiser got arrested.

      Thanks for that link. I just saw this interesting comment there from Jeff Mahoney at Suse Labs:

http://linux.wordpress.com/2006/09/27/suse-102-dit ching-reiserfs-as-it-default-fs
(excerpt)

ReiserFS v3 is a dead end. Hans has been pushing reiser4 for years now
and declared Reiser3 in maintenance mode. Any changes that aren't bug
fixes are met with violent resistance.

end quote

      I have some thoughts on the case of Nina Reiser on my site at:
Noted coder Hans Reiser arrested for wife's disappearance
http://www.justiceforchandra.com/forums/viewtopic. php?t=2899

  rd

Re:Old news

[Anonymous Coward]

If you are going to link to a site with pictures of dead toddlers being carried, you could at least note that the site is not safe. It's rather disturbing to find oneself having to explain this to your children on Sunday morning.

Re:Old news

[rbanffy]

Wait.

Did Nina Reiser want some non-bugfix changes on Reiser3 and met violent resistance?

never mind...

Re:Old news

[rbanffy]

Sorry, but I had to ask:

Was the decision announced before or after his ex-wife went missing?

You're forgetting

[joe_n_bloe]

4. The bitch set me up (Marion Barry)
5. The glove's too tight (OJ)
6. Is that Chewbacca here? (Chewbacca defense)

Re:You're forgetting

[Dausha]

"4. The bitch set me up (Marion Barry)"

I think this defense should be the one he goes with. I mean, it takes a sick bitch to kill herself then hide her body. She had to have planned it for weeks.

Seriously, though, how hard would it be to splatter one's own blood in your home and ex-husband's car, then flee the country?

Re:You're forgetting

[FusionDragon2099]

Is that Chewbacca here? (Chewbacca defense)

Turn in your nerd license, that's not how the defense works. Here's how:

"Ladies and gentlemen of the jury, this is Chewbacca. Chewbacca is a wookie from the planet Kashyyyk. But Chewbacca lives on the planet Endor. Now think about that; that does not make sense. Why would a wookie, an 8 foot tall wookie, want to live on Endor with a bunch of two foot tall ewoks? That does not make sense! But more importantly, you have to ask yourself, 'what does that have to do with this case?' Nothing. Ladies and Gentlemen, it has nothing to do with this case. It does not make sense!"

Re:You're forgetting

[Anonymous Coward]

It honestly doesn't matter. The Endor Holocaust: http://www.theforce.net/swtc/holocaust.html

Re:You're forgetting

[identity0]

7. "I'm just a patsy-OH NO I'VE BEEN SHOT!" (Oswald defence)
8. "You want the Truth? YOU CAN'T HANDLE THE TRUTH!!" (A Few Good Men defence)
9. "I'm telling you, it wasn't me! It was a one-armed man! You've got to believe me!" (The Fugitive defence)
10. "These are not the evidence you are lookiing for" (Obi-wan defense)
11. "That depends on what the definition of 'kill' is." (The Clinton defence)
12. "Putting this 'evidence' out for anyone to read is helping our terrorist enemies." (The Bush defence)

Re:You're forgetting

[Anonymous Coward]

13. I don't remember. (The Reagan defence)

arrest aside...

[Anonymous MadCoe]

It seems that ReiserFS really depends on 1 guy. For any company this is a risk. It sounds reasonable to me to stay away from products and features like that.

Re:arrest aside...

[Bronster]

Hans Reiser really hasn't been supporting Reiserfs (v3) for a while now, most of the work has been coming out of SuSE anyway. At least, the patches we've been running (mostly merged into mainline now) have been coming from them, and we're not even a customer.

I guess we won't be bugging Hans about issues for a bit now. Hope he's innocent and it gets resolved quickly - must suck for everyone involved right now.

Re:arrest aside...

[RAMMS EIN]

``It seems that ReiserFS really depends on 1 guy.''

What about all the paid developers working for Namesys?

Re:arrest aside...

[SumoRoti]

It seems that ReiserFS really depends on 1 guy. For any company this is a risk. It sounds reasonable to me to stay away from products and features like that.
Of course, if Linus Torvalds did zoophilia with penguins, I'd never use Linux... Or if Steve Jobs wears leopard panty...

Re:arrest aside...

[Anonymous MadCoe]

True, though the noise and discussions I've seen around the arrest suggest he _is_ the key person.
(I could be wrong, like I said it seems).

And Then the question is is one of those guys actually able to fill in that role.

Re:arrest aside...

[RAMMS EIN]

ReiserFS is open source, so it can be maintained as long as needed.

Re:arrest aside...

[TheRaven64]

It can be maintained as long as needed and someone is willing to pay for it. Someone capable of maintaining something like ReiserFS is not going to be cheap. If enough people value the development, then it will continue. If not (and if there isn't a filesystems guru who decides that they want to play with ReiserFS) then it will not be maintained. Since Linux is notorious for not having stable kernel APIs, any future kernel release could easily break it (although this is less likely to happen if any of the VFS guys use ReiserFS).

Previously, ReiserFS (version 3) was maintained primarily by a guy at SuSE (not by Hans, who claimed it was in 'maintenance mode' and was working on v4). He's now decided that it's not worth the effort. Maybe someone else will pick it up, maybe they won't.

Just rename it

[joe_n_bloe]

As I said in another post, if you want to market it, just rename it "The Filesystem that Serves You Right." ServesYouRightFS anyone? :-/

Re:Just rename it

[bernywork]

ReiserFS --> ServesYouRightFS...

Imagine the jokes if Hans was found guilty.... Maybe not.

Re:Just rename it

[kimvette]

I hope he is not guilty; I hope that his wife simply tried to pull a scam and is actually OK and is found hiding in Russia or somewhere. Even though it'd have been a scummy thing for her to do, at least their kids would not have lost their mother. That's the single worst part of the whole thing -- oh wait, am I saying "think of the children?" Shoot me now, please!

Re:Just rename it

[Killjoy_NL]

Thinking of the children is not a bad thing here.
I also hope she isn't dead and that his name is cleared, but if he did it, I hope they find evidence and that they convict him.
I'm not ruling out any possibilities.

Re:Just rename it

[spathi-wa]

I agree, mentioning children is not wrong when the children are already in the core picture

Re:Just rename it

[zaf]

You're right. The only safe solution is to rename the filesystem to a name that's already stood the test of a trial by jury and been found innocent. I hereby place my vote for renaming reiserfs to ojsimpsonfs.

Nothing todo with Hans' arrest.

[linuxpoweredtrekkie]

Several commenters appear to think that this is due to the arrest of Hans, In fact it was announced over a month ago, before any of the stories about Hans broke. The original announcement is from the 14th september http://lists.opensuse.org/opensuse-factory/2006-09 /msg00542.html

Re:Nothing todo with Hans' arrest.

[Rob Kaper]

Although "we" never noticed it, the Reiser news has been around for much longer, check Google News sorted by date. I read some older articles (prior to the arrest, prior to Nina going missing). Although the decision is defendable on its own, don't ignore the fact that reiserfs has been under control of a workaholic mathematician in a tough divorce while $170K in debt for a quite a while. This must have affected development and as such had an indirect influence. In the end it's all related.

Murder is not a socially acceptable problem though

[Anonymous Coward]

He's accused of murder. It's not like he stole a 6-pack from the corner 7-11.

Smart move, just a little late

[Anonymous Coward]

When it comes to performance between the filesystems (reiser vs. ext3 vs. xfs) then I don't have much to comment, but with regards to security... I've used reiser for quite some time but in the end threw it away because it just couldn't cope with what I wanted..

First your average backup. Yes, I'm well aware that you can always tools like tar but really.. Its the same deal with Sun's current development ZFS: it lacks the option to decently make a backup. Yes you can use tar, but I don't consider this decent. I'm talking about tools like backup/restore (ext3) or even native "ports" like xfsdump/xfsrestore. Easy, fast and reliable. Make a whole dump (or increamental), you can then either restore the whole session or use an interactive shell to merely grab the file(s) you're after. Naturally it also supports commandline parameters. And Reiser? IIRC (correct me if I'm wrong please) its even longer around than xfs, and even xfs managed to get me something decent for making backups...

Last but not least; crash recovery. I know, this is threading on thin ice since these results cannot be reproduced perse but the whole nature of reiser makes it good and bad for workstations (like SuSE). The good part is its speed, the way it caches and writes data in such a way where it tries to store things in one specific part makes it faster. I can't comment if reiser really is faster than others, I never noticed it. But the bad part is also that if you have a crash on your hands (just turn of your computer right now. No, not a shutdown but keep the powerbutton pressed untill it goes "poof") and reboot chances are very high that you just lost valuable data.

The theory behind journaling should give you some protection against this, and normally it does, but its my experience that whenever something like this happened on a box which was using reiser I lost just too many files. Several files in /etc used to become corrupt, binaries started going haywire and the worst part: because the index wasn't affected it was quite hard to detect these bad files.

Eventually I moved to XFS myself and never bothered looking back. Its not perfect, absolutely not since on XFS you too can experience situations like I just described. But in that same environment where I sometimes had to endure a powerloss I noticed that the frequency in which my data became corrupt was far and far less than with reiser. So my conclusion: reiser isn't the best when it comes to keeping your data safe. Its also a conclusion which has been backed up by other people who experiences the same problems in a more or lesser degree.

So my comment: finally Novell is coming to its senses. IMO they should have done this years ago, either going to XFS (my favorite) or ext3 where the latter is ofcourse the most logical choice considering how this evolved from ext2 (which, strangely enough, used to be the default on SuSE. I never did understand why they'd move away from it).

Re:Smart move, just a little late

[DRobson]

[...]the bad part is also that if you have a crash on your hands (just turn of your computer right now. No, not a shutdown but keep the powerbutton pressed untill it goes "poof") and reboot chances are very high that you just lost valuable data.

Dude, that's the case with everything, it's an inherent risk with caching. Journaling is not, as you implied, designed to fix this problem. Journaling ensures that the system can be rapidly recovered to a known good state. What you're looking for is some sort of flushing interval, and while I'm not up on the specifics of ReiserFS I'm sure it's a configurable parameter.

Re:Smart move, just a little late

[RAMMS EIN]

Let me start by saying I'm not disagreeing with you, just pointing out that different people may have different experiences (AKA YMMV).

``First your average backup. Yes, I'm well aware that you can always tools like tar but really.. Its the same deal with Sun's current development ZFS: it lacks the option to decently make a backup. Yes you can use tar, but I don't consider this decent. I'm talking about tools like backup/restore (ext3) or even native "ports" like xfsdump/xfsrestore. Easy, fast and reliable. Make a whole dump (or increamental), you can then either restore the whole session or use an interactive shell to merely grab the file(s) you're after. Naturally it also supports commandline parameters. And Reiser? IIRC (correct me if I'm wrong please) its even longer around than xfs, and even xfs managed to get me something decent for making backups...''

I believe backup tools that depend on the specifics of filesystems are a bad idea.

When you go looking for filesystem-independent backup tools, I'm sure you'll find plenty (the recent thread here on Slashdot may be a good starting point). I myself keep most of my data in Subversion repositories and databases; backups are made through the appropriate backup tools. Whatever is left on the filesystem is synchronized between a couple of computers using rsync.

``The good part is its speed, the way it caches and writes data in such a way where it tries to store things in one specific part makes it faster. I can't comment if reiser really is faster than others, I never noticed it.''

In the tests I ran, it wiped the floor with ext2 and (OpenBSD) ffs, especially when extracting lots of small files. I have no idea how it compares to more modern filesystems like XFS, ZFS, etc.

``But the bad part is also that if you have a crash on your hands (just turn of your computer right now. No, not a shutdown but keep the powerbutton pressed untill it goes "poof") and reboot chances are very high that you just lost valuable data.''

Although I have lost files on ReiserFS partitions, I've lost way more on ext2 and (especially) HFS+ partitions.

``The theory behind journaling should give you some protection against this, and normally it does, but its my experience that whenever something like this happened on a box which was using reiser I lost just too many files. Several files in /etc used to become corrupt, binaries started going haywire and the worst part: because the index wasn't affected it was quite hard to detect these bad files.''

Often when files seem to be missing after a crash, fsck has been able to recover them for me. This goes for ext2, reiserfs, ffs, and hfs+. Reiserfs is the only one of these on which I have never gotten the filesystem so broken it couldn't be fixed anymore.

In case people are wondering where I get my data from: I work with a lot of old hardware which sometimes fails, laptops that run out of battery or are dropped on the floor, accidentally unplugged power cables, and the occasional unclean shutdown.

Re:Smart move, just a little late

[Homology]

In the tests I ran, it wiped the floor with ext2 and (OpenBSD) ffs, especially when extracting lots of small files. I have no idea how it compares to more modern filesystems like XFS, ZFS, etc.

Did you enable "softdep" on the partitions you used (for OpenBSD)? This will give a big boost for some file operations involving many files (like untarring). Note that Linux by default enables async, which OpenBSD does not.

Re:Smart move, just a little late

[segedunum]

Its the same deal with Sun's current development ZFS: it lacks the option to decently make a backup. Yes you can use tar, but I don't consider this decent. I'm talking about tools like backup/restore (ext3) or even native "ports" like xfsdump/xfsrestore.

What? That's why you have LVM and snapshots. Am I missing something here? Backup features in the filesystem is generally a bad idea.

The good part is its speed, the way it caches and writes data in such a way where it tries to store things in one specific part makes it faster. I can't comment if reiser really is faster than others, I never noticed it.

Reiser totally wipes the floor with ext filesystems on just about any workload, especially on small files. Imagine a scenario where you have a fileserver that serves lots of Word documents etc. to people. Suse's customers sure are going to notice the difference. XFS blows ext away on large files, and don't run VMware files on an ext partition.

But the bad part is also that if you have a crash on your hands (just turn of your computer right now. No, not a shutdown but keep the powerbutton pressed untill it goes "poof")

I saw ludicrous posts like this many a time on Gentoo's forums. No filesystem will guarantee to save you from this.

But in that same environment where I sometimes had to endure a powerloss I noticed that the frequency in which my data became corrupt was far and far less than with reiser.

On XFS? I doubt it. It is recommended you use a UPS with XFS filesystems.

Re:Smart move, just a little late

[RAMMS EIN]

``Did you enable "softdep" on the partitions you used (for OpenBSD)?''

Yes. But thanks for the tip.

Re:Smart move, just a little late

[myowntrueself]

IIRC (correct me if I'm wrong please) its even longer around than xfs

In Linux, yes.
In general, no.

XFS came from SGI who had been using it on their systems for a very long time. Don't know the exact dates but I would guess it predates Linux.

XFS is good and reliable.

One thing though; when it does have problems, you better hope that the filesystem is small enough for the available memory to fsck.

When XFS gets over a few hundred gigs you need more than 4G of RAM to fsck it. No joke.

Re:Smart move, just a little late

[arth1]

And Reiser? IIRC (correct me if I'm wrong please) its even longer around than xfs, and even xfs managed to get me something decent for making backups...

Consider yourself corrected. XFS went production in '94, when IRIX 5.3 XFS was released (it had been installed on an adhoc basis for certain customers in '93).
ReiserFS came out in 2001.

Thins aren't looking up for Hans.

[dannycim]

Geez, now blood's found in his car, and with the passenger seat missing, history of abuse, guy is arrested with $8,900 and his passport on him...

If he were a famous football player, he'd have a chance, but I don't think a filesystem developer can muster up a "dream team".

I expect other distros will knee-jerk too.

$ mount /dev/hda3 on / type reiserfs (rw)

Re:Thins aren't looking up for Hans.

[Gothmolly]

Nothing to do with reiserfs, but that is a stunningly crappy article. Nina "kicked him out" ? What sort of language is that? Also, weird non sequiturs in the article make it difficult to determine what was imporant and what wasn't - "also included was a receipt for a syphon pump". Maybe he had a snickers bar wrapper in his pocket too? Ooooh!

Lastly, this just shows how you SHOULDNT buy stuff on credit cards or ATM cards, they pulled his records and found what books he bought.

Re:Thins aren't looking up for Hans.

[pipatron]

Anyone who owns a company where most of the developers are located in Russia would probably be prepared to travel on quite a short notice.

Can't explain the blood though. :)

I wouldn't think much of the "abuse history". Such accusations are supposedly very common in a divorce, especially when both parents wants to keep custody of the children.

Re:Thins aren't looking up for Hans.

[NDPTAL85]

What, everyone is supposed to use cash on the off chance that they might kill their spouse so that it will be harder for the police to figure out if you really did it or not?

Even for slashdot thats some fucked up logic.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

You're out of your tiny little mind, bitch.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

this just shows how you SHOULDNT buy stuff on credit cards or ATM cards, they pulled his records and found what books he bought.

He paid cash, we pulled it from the surveillance cameras.

By the way how are you enjoying that magazine you bought last Friday night? We saw you.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

Ask Slashdot: hi, I'm thinking of murdering my wife, what does everyone think is the best way of maximising my chances of avoiding detection?

Re:Thins aren't looking up for Hans.

[vidarh]

I have no basis for saying whether or not he's guilty, but take a bit of a closer look at the claims:

The article you link to say that "forensic tests on the blood cannot exclude Nina Reiser as a donor". Translation: They do not have a definitive DNA match, which means either that someone did a rather good job at cleaning, and/or that it's not new. The police better get more certain than that if they want to rely on it in court.

The only "history of abuse" I am aware of is a claim by Nina that he was abusive as part of their divorce battle. From what little I've seen of the case the claim was never substantiated, and the temporary restraining order was eventually dropped, though I may be mistaken on that.

As for having his passport on him, I very often have my passport on me. In my case it's simply because it's the only photo id I have (I don't have a drivers license) that is valid in the UK (where I live), and so it's handy to have on me. But more than that, because I'm a frequent traveller I generally keep my passport IN the jacket I usually use when I travel because I just forget it (I also frequently have months old stubs from boarding passes in my pockets...) or don't care about taking it out.

As for having lots of cash on him, there are plenty of valid reasons, many of which could also explain the passport. One example would be the need to quickly transfer money to pay his Russian employees salaries - I've more than once had to carry large amounts of cash to Western Union etc. to do rapid money transfers to foreign countries, or to buy a bankers draft for that matter, for perfectly legitimate reasons. At one point after I moved to the UK, for instance, I had to pay about 10.000 pounds as a security deposit and first few months rent to rent a flat with some colleagues, and ended up having to max out my debit card cash limits at the ATM's outside the bank branch because they wouldn't debit the money directly, and getting my Norwegian bank to transfer money would take several days.

The banking system is really inconvenient if you frequently deal with foreign countries, and cash often becomes the only viable means of shortening the time it takes.

Countries like Russia in particular are problematic. At one point I was paying a Russian guy for a license to some software he'd developed, and the money had to be transferred from my bank in Norway via a bank in New York, because my Norwegian bank just wouldn't do transfers to Russia at all due to fraud risks (it's not a general thing with Norwegian banks - it depends on which particular bank you deal with, and you'll find the same in the US), so I had the choice between opening an account at a different bank and carry cash (or waiting 3 days for a direct transfer), or pay extra for the transfer via New York. In that case I chose the latter, since the transfer was happening in dollars anyway, and it was a onetime thing.

I'm not saying any of these are right, but that it's a bit early to assume that they prove anything.

Re:Thins aren't looking up for Hans.

[vidarh]

If I buy a book about serial killers, and my wife happen to disappear a week later, do you think it would be better or worse for me if the police found the purchase on my credit card statement?

That is one of many reasons is why you should think about what tracks you are leaving, even when there is nothing nefarious about your purchases.

In this case there is at least one possible explanation: If I was a likely murder suspect I damn well would like to know what to expect from the police.

There are of course reasons why he might have done this that aren't quite as nice. In either case I must say I think it was a rather naive and stupid move.

Re:Thins aren't looking up for Hans.

[rolfwind]

One shouldn't have to explain the reason for having cash on their person. Europeans regularly have more cash on them because it makes them feel more secure while in America it's normal to run to the ATM before buying a $10 Pizza, unless they use their credit card.

It's just a cultural difference. Maybe it mainly stems from the fact that here the police can consider you a drug pusher if you have $X,000 on you and confiscate said monies without any other proof. Then you have to prove it was yours to get it back. You know, guilty until proven innocent.

Re:Thins aren't looking up for Hans.

[NDPTAL85]

You can always just go to a library and read it there without ever checking it out and no one would know you ever read it.

Re:Thins aren't looking up for Hans.

[jb.hl.com]

99% of predicted replies: "Use Linux".

Re:Thins aren't looking up for Hans.

[Bastard of Subhumani]

If I buy a book about serial killers, and my wife happen to disappear a week later, do you think it would be better or worse for me if the police found the purchase on my credit card statement?

Better - you could use the "If I was going to do it I'd have been more careful to cover my tracks" defence.

Re:Thins aren't looking up for Hans.

[evilviper]

What, everyone is supposed to use cash on the off chance that they might kill their spouse so that it will be harder for the police to figure out if you really did it or not?

Do you really think they wait until they're convinced someone committed a capital crime before people look through their CC records?

They'll look through them just because it's TUESDAY.

Cash is MUCH harder to trace, accepted everywhere, cheaper, and FASTER (no need to show ID and sign papers to buy a pack of gum).

Re:Thins aren't looking up for Hans.

[TheRaven64]

Cash is faster? You've got to be kidding. For a cash transaction:

1. Make sure I've gone to an ATM in advance and got enough cash for the transaction.
2. Find something approximating exact change.
3. Wait for cashier to count it.
4. Wait for cashier to hand me my change.
5. Count the change.

For a credit card:

1. Hand over card.
2. Enter four digit pin.

Even step one of the cash transaction is slower than the entire credit card transaction.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

> Nothing to do with reiserfs, but that is a stunningly crappy article. Nina "kicked him out" ? What sort of language is that?

How about "Nina got a permanent restraining order against him".

Sounds a little bit like she gave him the boot.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

Maybe it mainly stems from the fact that here the police can consider you a drug pusher if you have $X,000 on you and confiscate said monies without any other proof.

I bet you can't cite an example of that happening without the police having some valid suspicion (e.g., someone they've arrested before, or a neighborhood known for pushing drugs).

Re:Thins aren't looking up for Hans.

[evilviper]

Make sure I've gone to an ATM in advance and got enough cash for the transaction.

This one is absolutely ridiculous. I could just as easily include "Looking around the house for my credit card," "Going to the ATM to check my debit card balance," or any other such nonsense.

Find something approximating exact change. [...] Count the change.

Well, I guess if you can't count, that might take several seconds. Even if so, it probably only takes a couple seconds longer than pulling your credit card out and putting it back.

Wait for cashier to count it. Wait for cashier to hand me my change.

If they've been at the job for any length of time, they can do all this in no time at all.

This is all NOTHING compared to handing over your ID and card, waiting while the cashier compares them, signing your name on the paperwork, writting the exact charge down in your notebook, etc. Never mind having to scan the card repeatedly, waiting for the approval message to come through, etc.

You can make up all the senarios you want, but we've ALL be in line behind people with credit cards, and people with cash, and have all seen those with cash flying through the lines, while those with credit cards hold it up for MINUTES at a time. It's just not even debatable which is faster.

Re:Thins aren't looking up for Hans.

[rolfwind]

Oh, I know it happening to friends who didn't know better, but if you want a cited case, I can give you enough without looking to far on google, here's one:

http://www.aclu.org/police/forfeit/14599prs2000112 1.html

Re:Thins aren't looking up for Hans.

[TheRaven64]

I don't know what kind of backwards country you live in, you you really should get them to upgrade their banking system. First, how do you get cash if not from an ATM? These days pretty much everyone is paid by standing order, and credit cards can be paid by direct debit so the money requires no intervention on your part to go from your employer to the credit card company.

Why on earth would I hand over ID to make a credit card purchase? Why would I sign anything? Signatures haven't been valid for credit card payments for a while now (credit card companies accept no liability for fraud if the merchant accepts a signature, only if they enter their PIN), and entering the PIN takes seconds.

Writing the charge down in a notebook? Now your grasping at straws. I can get a credit card statement and a list of recent transactions online (and download this in CSV format) from my card issuer, and if I want to check for errors I just pick up the receipt at the till.

Re:Thins aren't looking up for Hans.

[Moridineas]

Good point, but he bought the books AFTER his ex disappeared.

Re:Thins aren't looking up for Hans.

[djdavetrouble]

Its like the plot of a noir movie. Dude macks open source filesystem visionary's wife, plies her with ecstasy (MDMA), turns her out into some bdsm freak (you saw that detail didn't you), only doesn't realize that the open source genius is a little off:

Hans Reiser purchased two books "Homicide'' by David Simon and "Masterpieces of Murder" by Jonathan Goodman on Sept. 8 from Barnes & Noble in Berkeley.

and is capable of ...... murder ...

its not looking pretty, or this is the setup of the century.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

He did? Or someone who wants to get rid of him did? Were the books found? Were they read? Would someone who knows about technology buy such knowing they are suspected of murdering their wife? After she has been murdered? Or was she still alive when he bought the books?

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

Usually people use cash for small amts of payment and a creditcard/debitcard/valueloadcard for large amts. For a pack of gums you'd rather give a dollar and begone, give the guy a tip or buy 2 packs. Doesn't cost you a transaction. It costed you money when you withdrew it from ATM but you used that money at various places for small things to buy. Quick and easy indeed.

The problem with a creditcard is that you create debt the moment you use it. The problem with citizens in debt you may or may not know. Goes beyond scope. Has to be said here though. A debitcard/valueloadcard, or ATM card which can't 'go red', is much more like having cash at hand.

Cash will (or does) contain RFID chips btw. Easy traceable then.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

She is his wife; not his ex. Assuming he has never divorced someone he has no ex, legally speaking. His wife is not dead either, officially. His wife is missing. So far the anal part...

What you state is true but doesn't mean anything though. Assume (!) he made her disappear and assume (!) he murdered her (which may or may not have happened both in the same timespan of say a few hours, but could have happened much longer; such would be found out if she'd be found dead). If he bought the books after she had been murdered, one could argue he had planned bad. E.g. that the murder wasn't planned. It went out of hand. If he bought the books before she had been murdered which may be after she disappeared it could be argued he planned well.

This is worse case scenario thinking.

It could also very well be the case that he bought the books because he suspected he'd be suspect of murder. If you suspect you will be suspected for murder before the victim has been disappeared, that is suspicious to say the least. It doesn't prove a correlation between the buying of the books and her disappearing / alleged being murdered though. For cops this would be reason for interrogation since it is a clue to them suspecting you, and then they'd like to crack you on a whip (but they cannot prove (yet)). Here, that is not the case though as he bought them afterwards after the disappearing (_but_ we don't know when/if she got murdered).

It doesn't matter (for now) though due to the fact that the books themselves are no evidence of the murder. There is no proof for either theory I stated hence they are irrelevant (for now). Or rather, should be irrelevant. Important facts are missing. Only usage of this is for the media, they love blood. This guy is most likely already convicted for life because of the missing dots and the media leaking (for now irrelevant) facts.

PS: We know his motive, such is pretty clear, however given he hates the other guy more it is surprising he has not been murdered by him as well. What about his alibi? If he adds code to CVS, and he programmed that day, such may be proof. Basically given he uses computers such traces can work in his advantage. Even if he only e-mailed aunt Tilly whole day that is interesting for defence against alleged alibi.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

MDMA and psychoactive drugs can be used for mind control, as can torture (SM could be a form of torture IMO), as can media. There is scientific proof for this.

Any human is capable of murder. A cop is capable of murdering. Do you suspect them first in case of murder? No, you don't. He has a motive indeed if that is what you meant.

He purchased the books after she disappeared. We do not know if/when she got murdered.

Re:Thins aren't looking up for Hans.

[Schraegstrichpunkt]

0.9% of predicted replies: "WTF are you asking Slashdot? Hire yourself a real assassin."

Re:Thins aren't looking up for Hans.

[Gojira Shipi-Taro]

There's also the whole thing about it being easier to rob someone who's carrying cash. I know I don't carry more than $20-30 as a rule (usually nothing at all, unless I'm planning on dining somewhere that I don't want the serving staff to access my credit card number).

Not carrying cash also makes interactions with beggars a hell of a lot easier. "Sir do you have a dollar?" "Nope, don't carry cash. Sorry." Not that I would have given the meth addict in front of the grociery any money anyhow.

I wouldn't carry the kind of cash Hans was found with without a carry permit, and a very deadly concealed piece.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

OT. Just curious, where did you get your debit card? Is it for UK citizens only, or also for EU citizens? I am a EU citizen and am searching for a debit card. So far I've only been able to find debit cards for US and UK citizens.

PS: "to assume that they prove" is a contradictio interminus. You cannot assume to prove. You prove true or false. When you assume, by definition, you are not sure wether it is true or false. If done well, you try to prove your assumption. The evidence gathered so far is thin air to prove, I agree w/you on that.

Re:Thins aren't looking up for Hans.

[dhasenan]

Murder suspects shouldn't have to explain the reason for having cash on their person.

Sounds a bit different in that light, no?

Re:Thins aren't looking up for Hans.

[Reservoir Penguin]

I believe there is a conspiracy against OSS developers. First Alan Cox is almost killed by and explosive charge hidden in his laptop. Now Reiser is set up. What next? Linus has a few beers and dies in a car crash? RMS found hanged in his room?

Re:Thins aren't looking up for Hans.

[tigersha]

... and connect it to some weird Lego Mindstorm Killbot that you copied off a design for a Romulan Android Warrior you once saw in the Trek Tech Manual.

Re:Thins aren't looking up for Hans.

[AaronLawrence]

If certain large software companies thought there was a serious threat from OSS, I'm sure some of those billions could accidentally be diverted to getting rid of some of the key players.

Silly conspiracy theory ...

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

Eh? My credit card doesn't even *have* a PIN. I have never once been prompted to enter a PIN. My credit card cannot be used at an ATM to withdraw cash -- it has no PIN. I have all purchases I've made, with the exclusion of those at Walgreens that are under $15 or so. And they didn't ask for a PIN either.

I have never heard of anyone being asked for a PIN for a credit card. Methinks you just have a debit card and are too stupid to realize it.

Re:Thins aren't looking up for Hans.

[einsteinx2]

Last 0.1% of replies: F1RST P0ST!!!!111

Re:Thins aren't looking up for Hans.

[TheRaven64]

Walgreens? I guess that means you're in America. In that case your banking system is about 20 years behind the rest of the world (well, 30-40 years behind places like Japan). I absolutely loath having to have any interaction with my US bank since it's like stepping back into the dark ages.

Re:Thins aren't looking up for Hans.

[Anonymous Coward]

It is incredibly easy for women to get restraining orders against ex-partners. There is no real proof required at all.

In fact many sleazy types recommend it for women about to undergo divorce proceedings, because it becomes a sort of proof itself, to help them get better settlements.

A restraining order without knowing the actual reasons it was granted, or any evidence of abuse, means absolutely nothing in this day and age.

Re:Murder is not a socially acceptable problem tho

[mustafap]

>Murder is not a socially acceptable problem though

He remains innocent until proven guilty by a jury, unless you know better.

Think of the women!

[Analein]

http://ns2.reboot.net.au/~uridium/reiser.png

Re:Think of the women!

[Polly_Morf]

Now whats so funny about Slashdot users go away!!!1 this is a tiny home adsl connection. Avast! ?? :D

Re:Think of the women!

[KDR_11k]

"Object not found"

Did you link to his wife?

Re:Think of the women!

[Anonymous Coward]

Nah.. I moved the object. I only have a 6.8kb/sec upstream and 700,000+ hits = pain. Go try google cache their images. Soz. 256/64k adsl = she cannah handle teh speed captain!!!1

Mirror:

                          http://lutin.jard.in/~dek/reiser.png

Is the charge worth getting rid of a product?

[Skuggi]

Come on now, just cause someones charged with murder, should that really stop the progress of technology? I can see changing the name if he's convicted, but what happened to innocent until proven guilty...

Re:Is the charge worth getting rid of a product?

[AusIV]

As several other people on this thread have noted, Reiser3, which is what Novell is dropping support for, has been maintained solely by Novell for a few years. Hans Reiser dropped support for Reiser3 quite a while back, insisting people should use Reiser4. Novell didn't want to move to Reiser4, and that would be an especially shakey move with recent events, so instead they chose to go with the more stable, slightly slower Ext3 file system. This was a decision in the works before Reiser's arrest.

XFS anyone?

[Thaidog]

Why not move it to XFS? This says "commercial production ready" to me.

Re:Murder is not a socially acceptable problem tho

[Anonymous Coward]

I give you the word "society", if thats spelled right. Anyways, it depends on what society you are in, weather or not killing someone is allowed, and under what circumstances that it is allowed in. For example, wayway back, ritual fights to the death where accepted, now days, only if its court ordered, and its not much of a fight with one of em being tied down and all, which is why they call it executions.

Since its his wife, you could say that he couldent have possibly killed her. Since they where married, they could be seen as one, thus, if he did do this, he dident murder her, but he commited suicide, but since hes still alive, we can only deduce that he only killed part of himself, thats call this process, divorice.

Anymore questions?

ext3 more reliable? Whatthe!

[mcbridematt]

it claims ext3 is more stable and will 'soon' match performance with the newer ReiserFS 4.

Gee, ext3 must've matured a lot in the past few years. I stopped using extX filesystems long ago because they lost files after power cuts waay too easily. ( I could bork an old RedHat install simply by pulling the plug/rebooting several times ). Moved to reiser then xfs and barely lost anything if I had to force a reboot.

Re:ext3 more reliable? Whatthe!

[joe_n_bloe]

You didn't have a UPS that could last through an install? Or you were just showing off?

Re:ext3 more reliable? Whatthe!

[Anonymous Coward]

Completely wrong. ext2/3 have demonstrably more reliablity in the face of shitty hardware/power outages. It's been shown many, many times. They were developed for it... unlike basically any of Reiser's filesystems... which all assume perfect hardware and conditions, and hence have a documented history of shitting on file systems.

Re:ext3 more reliable? Whatthe!

[asuffield]

I stopped using extX filesystems long ago because they lost files after power cuts waay too easily.

That's still better than reiserfs, which does not need a power cut in order to lose data. I still recall a comment from a tech support area I used to frequent: "reiserfs runs really fast until it crashes and you lose all your data. As a result it has a lot of ex-users who are now sadder but wiser."

It is also important to remember that ext3 can be configured for a number of different points along the speed/safety tradeoff, so any stories about problems (with speed *or* safety) should state which mode they were using.

Re:ext3 more reliable? Whatthe!

[joto]

Gee, it must have been a long time since you were using ext3. By the time ext3 was announced, it was a minor modification to an already very mature file-system.

Remember, ext2 was *the* file-system the vast majority of linux users all over the world used. ext2 survived through Yggdrasil, Slackware, and RedHat. Then people started porting various journalled file-systems to linux, and ext3 was created as a journalled variation of ext2. Even by the first version, ext3 was probably safer than XFS or JFS, who had to be modified extensively to fit inside the linux kernel.

It didn't take long for ext3 to take the place of the default file-system on linux, with ext2 having second seat. Other file-systems, such as reiserfs, XFS, JFS, and so on, hasn't got nearly the same amount of testing. Calling ext3 immature is like calling FAT untested.

Re:ext3 more reliable? Whatthe!

[Frumious Wombat]

I'm glad to hear somebody speak up for XFS, since though I was less than fond of SGI in general (greasy salescritters and insecure OS), I never had a problem with XFS. I used that FS, and never lost data, on machines ranging from Indigos (which I had to tape over the power button to keep users from shutting them off like a PC), through Origin2000s. Same story for JFS on IBM hardware from RS/6000 320H through SP2. This was at locations with dirty power, where sometimes the workstations would get powered down unexpectedly. (once I got to see the cause; it's an amazing sight when you see a squirrel electrocute itself on a transformer, and go up in a ball of flame)

Therefore, I have to politely ask, "What the *BLEEP* have you Linux people *DONE* to those filesystems?"

Re:ext3 more reliable? Whatthe!

[Jeff DeMaagd]

Yeah, a good UPS is like $50 now. If you are ever working on data worth more than $50 in terms of cost to re-collect or re-do if something happened to the power while working on it, then it basically pays for itself for the first power outage. For me, they are pretty good at preventing other electrical damage from ruining the computer too, in my experience, better than just using a surge strip.

It's not a substitute for a proper backup regimen, but it should keep you from losing a few hours worth of work.

Re:ext3 more reliable? Whatthe!

[Shawn is an Asshole]

I used to use ReiserFS3 and had too many problems with it. Several times I have files just disappear. Other times I had files and folders that not even root could delete. I had many files get mysteriously corrupted. The second hard drive developed some bad sectors and I ended up losing everything on that drive.

What I use now is EXT3 (with dir_index) for / and XFS for /home and also file directories for my servers. I have not had any issues with corruption nor any problems with files disappearing.

This was a while ago, though, when it was recently added to the kernel. The problems I encountered may have been fixed since then.

Re:ext3 more reliable? Whatthe!

[segedunum]

Remember, ext2 was *the* file-system the vast majority of linux users all over the world used.

Yer. And people were more than glad to get away from it. Truly awful performance, and not all that reliable on all the things people say ReiserFS fails on i.e. yanking the power cord out!

Other file-systems, such as reiserfs, XFS, JFS, and so on, hasn't got nearly the same amount of testing.

Reiser 3 has had years of testing in many environments.

Calling ext3 immature is like calling FAT untested.

Nobody particularly likes FAT either ;-).

Re:ext3 more reliable? Whatthe!

[segedunum]

Therefore, I have to politely ask, "What the *BLEEP* have you Linux people *DONE* to those filesystems?"

It's very simple. Many people involved with Linux, for political reasons, don't want to accept that ext sucks ass as a filesystem in many circumstances, which is why we get all this "Oh, you'll lose data if you don't use ext3!" comments. The net result is that good filesystems like XFS and JFS have been marginalised and haven't really been improved as much as they should. JFS is a particularly good filesystem that people like Novell should be looking at.

Re:ext3 more reliable? Whatthe!

[segedunum]

That's still better than reiserfs, which does not need a power cut in order to lose data.

For those of us who've had Reiser filesystems up and running for years, you might want to enlighten us as to how a Reiser filesystem can be screwed without something like a power cut (assuming no UPs etc.)

reiserfs runs really fast until it crashes

How does a filesystem 'crash', exactly?

Excuse me for being dubious about these types of comments, but I've seen them made for years and they still don't make any sense whatsoever to me - because they're absolute tosh.

Re:ext3 more reliable? Whatthe!

[evilviper]

Gee, ext3 must've matured a lot in the past few years. I stopped using extX filesystems long ago because they lost files after power cuts waay too easily.

Don't compare ext3 with ext2.

Non-journaling, ext2 was terribly unreliable... And even with that, it's not that the filesystem was unstable, just that it couldn't maintain integrity when mounted async. Sadly, for performance, it was mounted async by default, and few people either understood, or had enough good sense to change it.

( I could bork an old RedHat install simply by pulling the plug/rebooting several times ). Moved to reiser then xfs and barely lost anything if I had to force a reboot.

When I was first starting with Linux (Redhad 5.x), I lost large Ext2 filesystems on multiple occasions. With ext3, however, I haven't lost one yet.

The only problem I ever had was that fsck.ext3 didn't fix one filesystem error I was seeing (after a crash), but manually running fsck.ext2 on it solved the problem...

Re:ext3 more reliable? Whatthe!

[Anonymous Coward]

I had a drive that I abused the hell out of fail after about 3 years. Compiling and updating gentoo frequently, 2 gigabyte subversion tree of mostly small files, huge vmware images, you get the idea. It actually failed while reading date from a folder I hadn't used in a year or so. Linux put the fs in read-only mode after the error and I was able to make a backup of all good data over the network.

Then I rebooted for the heck of it and reiserfs ran its disk check for a couple hours. The warnings and errors just spewed up the screen the entire time. I had to reboot several times and continue the disk check. But in the end I had a usable system even on a disk that was actually failing. I was impressed. Any system that can recover from actual media read errors must be pretty good.

The moral of this story is, if you want to keep your data in good shape do s "find / | xargs cat" every once in a while so the drive can relocate or rewrite weak blocks.

Re:ext3 more reliable? Whatthe!

[joto]

Nobody particularly likes FAT either ;-).

FAT is probably the single most popular filesystem in the universe. At least these features makes it vastly popular:

• It's the only filesystem that works with DOS
• It is very suitable for floppies and for other small filesystems
• It is reasonably effective with small cache sizes
• It is universally supported (I doubt there even exist an OS with more than 100 users that doesn't support FAT in some form)
• There are no permissions, resource forks, quotas, or other things that complicates file-exchange between different systems
• It is simple enough for mere humans to recover data in a hex-editor, or write programs for it

Hopefully we are now entering an era when FAT can finally be laid to rest, with other filesystems such as UDF taking over. However, most flash devices are still using FAT.

Re:ext3 more reliable? Whatthe!

[Anonymous Coward]

Of course ext3 is more reliable. I heard stories that Reiser can kill your data (although it wouldn't admit it though) ...

Re:ext3 more reliable? Whatthe!

[asuffield]

How does a filesystem 'crash', exactly?

Exactly the same way as any other piece of software. Did you think filesystem drivers were somehow different to any other kind of code?

Re:ext3 more reliable? Whatthe!

[digidave]

ReiserFS caches a lot of data instead of writing it to disk. If your system crashes before it's written you will lose that data.

Re:ext3 more reliable? Whatthe!

[Gojira Shipi-Taro]

Well speaking from my own experience, I've never lost an entire filesystem to a single glitch,except for the time I trusted my files to ReiserFS. The provided recovery tool/plugin to fsck destroyed what was left of the volume. Of course any question to any forum that Hans and his cronies had access to was met with "we'll answer your question for $25, but the answer just might be "you're hosed".

I'll stick with ext3 (and eventually ext4), where the developers are reasonable, sane, and not alleged murderers. I considered Hans an arrogant fuck after my first interaction with him. This situation doesn't suprise me a bit.

Re:ext3 more reliable? Whatthe!

[SanityInAnarchy]

Don't compare ext3 with ext2.

While we're playing the ancedote game, I've had at least as many fscks under ext3 (yes 3, not 2) as under reiser4, not counting my overclocked system. But my overclocked system isn't stable anyway.

Re:ext3 more reliable? Whatthe!

[cloudmaster]

Speaking from my own experience, I've only lost files to 1) bad hardware and 2) extN file systems. Reiser's tool has always fixed any problem, except for on actual failed drives. The people I've heard problems from with Reiser typically have "other" problems with "difficult to learn" tools, systems, etc. IE, they're not exactly the sharpest kinives in the drawer. Kinda like this comment - from someone who had unidentified problems the "ony" time reiserFS was used. Huh, sounds like *lots* of experience... Sigh.

I will say, though, that ReiserFS is good at brinigng to light hardware problems that ext3 happily ignores. Then again, I personally *like* to know that things aren't working right, instead of waiting until it's too late.

Re:ext3 more reliable? Whatthe!

[evilviper]

My comment really wasn't about anecdotes.

Ext2 has well-known and understood issues (async mounted), which Ext3 eliminates. So, comparing the two is just simply a mistake.

anyone gets to shine a torch

[nurb432]

Sure we do, its part of human nature to poke fun at others misery. It doesnt have to be 'personal'.

And then

[Konster]

What's going to happen is this.

He will go to jail for a very long time and his file system will remain mostly pointless for a very long time, then it will go from being pointless to forgotten. In the end, Linux will still be a very bad OS and an incoherent one at that. Then in the year 2035 Firfox grabbed another 1% market share versus IE at 97%...and the gnomes made happy. The Linux mods dropped a mod bomb on my ass in order to shut me up. Conspiracy theorie abound until I was photographd in Monaco with Mike Jackson and at least three of us were seen wearing high heels purchased in Milan on a Tuesday.

Re:And then

[arth1]

He will go to jail for a very long time

If found guilty, probably jail for a very long time followed by execution. California has capital punishment.

Regards,
--
*Art

Re:And then

[Gothmolly]

For a white guy? I think not.

Re:And then

[Anonymous Coward]

You mean a white guy who offed some mail-order Russian prozzie?

Re:And then

[Overly Critical Guy]

Right, the state where O.J. got away with it for being black is racist against blacks.

Re:And then

[Anonymous Coward]

| If found guilty, probably jail for a very long time followed by execution.
| California has capital punishment.

chmod -x hans ...don't you guys know anything about linux?

-t

Re:And then

[arth1]

For a white guy? I think not.

California is the state that has the highest proportion of white guys on death row, with almost 40% of the inmates being white, 35% being black, and 19% hispanic.

Did I just hear the sound of a foot being extracted from a mouth?

Re:And then

[Gothmolly]

And when was the last time any of them actually died?

I think I hear the sound of your head being extracted from your ass.

Re:And then

[arth1]

And when was the last time any of them actually died?

At least[1] four out of the latest six executions in California have been of white men:

Darrell K. "Young Elk" Rich - white
Robert L. Massie - white
Stephen W. Anderson - white
Donald J. Beardslee - white
Stanley "Tookie" Williams - black
Clarence Ray Allen - other[1]

[1]: Does a white chocktaw classify as "white"?
http://a.abclocal.go.com/images/kgo/cms_exf_2005/n ews/clarence_ray_allen_121305_lg.jpg

I think I hear the sound of your head being extracted from your ass.

To think before you talk is like wiping your ass before you shit.
-- Arne Anka

Roumors

[El Lobo]

Roumors has that there are a coupl of pedophils and 5-6 thieves between ext3 developers as well... so what's next for Linux? NFS as native system?

this FP for GNAA!

[Anonymous Coward]

addresses w1ll Baby...don't fear

Re:this FP for GNAA!

[lennyhell]

yuo fgt. no rly f u. go away. k tknx.

ext3 Performance Matches Reiser4?!

[RAMMS EIN]

``ext3 is more stable and will 'soon' match performance with the newer ReiserFS 4''

Huh? In whose benchmarks? What about space usage? What about plugins for arbitrary attributes?

Re:ext3 Performance Matches Reiser4?!

[diegocgteleline.es]

What about plugins for arbitrary attributes?

You know, many people thinks that this "plugin-based FS" idea is just pure and old overengineering.

The main reason why people doesn't even dares to look at reiser 4 is that it's awfully *complex* - and that's quite an achievement for filesystems, which are all of them already quite complex.

Yes, reiser4 is damn fast. I'm not surprised, since they had the invaluable oportunity of writting a filesystem from the scratch and they're already linux fs experts. I think that most of kernel hackers think, even if they don't say it, "we can build a filesystem that matchs reiser4 and that it actually can be understood, read or maintained"

Re:ext3 Performance Matches Reiser4?!

[netdur]

why do you use ms word to post comment on /.?

Re:ext3 Performance Matches Reiser4?!

[Omnifarious]

Perhaps. But the single most valuable thing about both Reiser filesystems is how well they handle large numbers of small files. I hate Berkeley DB and its ilk with a passion. They take all kinds of valuable data that should be addressable with standard tools and obscure it in some weird format that I can't make any sense of without some specialized set of tools. Not only that, but they're slow!

I want to stop using these awful things. I want to use a hierarchical naming scheme to address the individual bits of data I'm stuffing into the filesystem without having to resort to stupid tricks with splitting up the name so I don't have anymore than 256 entries per directory.

None of the filesystems made for Linux aside from ReiserFS seem to even acknolwedge that this problem is worth solving. Personally, I think it is a major, short-sighted bellybutton gazing failure. The excuse seems to be "Well, you're using the filesystem in a strange way that nobody uses it in, so stop doing that!". But that's a completely circular argument. I simply do not WANT to contort my programs in such a ridiculous way to accomodate the failings of filesystem designs.

ReiserFS is fast and flexible. I've never had any data loss with 3. At least, not in the last 3 years or so. And I have a machine that will (for reasons of a bad motherboard) randomly lock up if I'm using both the disk and the Ethernet card heavily.

I don't really care that much about plug-ins. They're kind of a neat idea for having super-efficient storage for caches and stuff, but really I just want to be able to independently address millions of small pieces of data and have it be reasonably efficient.

Re:ext3 Performance Matches Reiser4?!

[cryptoluddite]

What you are looking for is probably the mount command. Seriously just look at the list of FUSE filesystems to get an idea of what can be done. We don't need one filesystem that is the best at everything, in fact that's impossible.

Even Berkely DB, which I also hate, would not be so bad if you could "mount -t berkdb -rw ./database.db /mnt/database".

Re:ext3 Performance Matches Reiser4?!

[oojah]

Even Berkely DB, which I also hate, would not be so bad if you could "mount -t berkdb -rw ./database.db /mnt/database".

Hmm, fun. I can't see a generic solution being really possible, but you could certainly do it relatively easily for a database format fixed at compile time. I don't like Berkeley DB either and wrote a simple tool for converting it to CSV which works in that way.

FUSE is actually great fun to play around with.

Re:ext3 Performance Matches Reiser4?!

[MrNemesis]

Here's a thought - would it be possible to create a FUSE module that would be able to present BDB file internals as filesystem objects? True, it'd be slow but at leas you'd have the advantage of being able to directly read and manipulate BDB info. You could probably do the same thing for SQL.

Heck, FUSE might even make it possible to embedded a ReiserFS partition as a file. You'd still have to go through the host filesystem, mind.

Re:ext3 Performance Matches Reiser4?!

[Omnifarious]

FUSE is an interesting idea, and may help to some extent (no pun intended). But, for example, I want to create a large database in a ~/.dir that consists of many mostly tiny files named after base32 representations of hashes of other bits of data. I could stuff all this data into a database, but the format of the data in the files is going to be highly variable, and some of them will be quite large, so I don't think that's a good solution unless blobs (which are pretty icky in and of themselves) are used.

OTOH, forcing people to mount ~/.dir as a FUSE filesystem to get decent performance doesn't seem right either. Though maybe it's a good idea because much of that data will be highly sensitive cryptographic information, and I could include encryption.

Hans Reiser

[alewar]

some thoughts about him

Re:Hans Reiser

[RAMMS EIN]

FTFB: ``Now, such smart guy like him, couldn't see it coming back in 1999?''

Well, just because he's smart doesn't mean he's good at judging people, or even scams. I know lots of smart people, and plenty of them have been scammed, cheated upon, etc. in cases where you'd think it was too obvious. That includes myself, by the way.

Re:Hans Reiser

[Anonymous Coward]

Good one. It boggles my mind how sometimes people cannot sense malice even when it is very obvious.

Re:Hans Reiser

[anagama]

Infatuation has a blinding effect on people. It's only by the time she's sucking $8000/month alimony out of him that he takes a moment to fo the math and relizes that visiting Nevada every weekend would be cost less and be more fun.

Re:Hans Reiser

[jsrlepage]

I am very weirdly in agreement with you. :S

Re:Hans Reiser

[ralphdaugherty]

she's sucking $8000/month alimony out of him... nbc11.com

      This is weird reporting. From the www.nbc11.com article Hans Reiser's Software Could Be Phased Out:

According to sources close to the investigation, Reiser had to pay $8,000 in alimony a month to Nina Reiser.

Sources also told NBC11 that Hans Reiser was seen walking around town with several thousand dollars in cash and his passport.

end quote

      The divorce wasn't final. Do you pay alimony while you're separated? Other articles report that he was to share child expenses with his estranged wife, and that she filed with the court that he hadn't paid it. The exact amount was not reported, but it wasn't alimony.

      The $8,000 per month, if it actaully is based on anything true, could be all the monthly bills that the family had, including the mortgage of their home that he no longer lived in, but even with that $8,000 is extreme.

      Also, how does one see someone walking around town with several thousand dollars and a passport? Were they sticking out of his pocket protector?

  rd

Hard to Believe

[countach]

Not that I've got my finger right on the pulse of FS development, but I find it hard to believe that ext3 is soon going to equal Reiserfs for all cases. Perhaps for a typical case, but ReiserFS was supposed to allow a lot of stuff that was not feasible with ext3 like efficiently having really small files, using the FS as a database, and a lot of other potentially groundbreaking research and abilities. I hope none of the good ideas get lost.

Re:Hard to Believe

[booch]

Yeah, I think most people here are missing the biggest impact Reiser was trying to make in UNIX file systems: semantics. First, he was trying to make small files efficient, so that each file could hold a smaller amount of data, more like a single field within a database row. So instead of having to parse /etc/passwd, you would put each piece of info in a separate file -- for example /etc/user/booch/password would contain the hashed password, and /etc/user/booch/shell would contain the user's shell.

But most importantly, he was trying to create a sort of extended attributes system whereby every file contains "sub-files". There could be a subfile containing ACLs for the file, a subfile containing quota info, subfiles for permissions, etc. And in addition to being able to find a file based on its filename and path, you'd be able to look up a file based on other attributes, like how you can SELECT a row from a database using arbitrary keys. The idea is to leverage the past several decades of what we've learned about databases to use in managing files. Because a filesystem is really just a database of blobs, indexed by pathnames.

Re:Hard to Believe

[psyclone]

Interesting use of lots of small files (with few lines) to store information, versus a few files (with lots of lines) to store information.

Also, a filesystem IS a database already. It is a hierarchical database, not a relational database.

Symptoms and Causes

[caudron]

What implications will this have for SuSE users, and ReiserFS's future as a whole?

Assuming this wasn't a rhetorical question, I'd say the answer is that the ReiserFS will be impacted only slightly by Novell's decision. The far bigger impact will be from a criminal conviction. Free Software is about community and community is all about those subjective intangibles like reputation, "coolness", and mob effects.

Whether we like it or not, this highlights a serious problem with the development model. Likewise, it indirectly highlights one of its strengths. Free software programmers are very much pack animals, like the rest of us. We tend to folllow the herd (I don't mean that in the modern "bad" sense of the phrase. We stick with those we know and enjoy hanging with. We do things we perceive subjectively as fun or cool. We join projects that interest us, we leave projects that offend or dissappoint or bore us. With the GPL, a company that relies on a no-longer-cool project can always pick up the banner and try to reinvigorate interest, but in the end the projects that have momentum have it because they have that special unnamable something that brings people to the fold.

Ubuntu got the right press from the right people at the right time. Is it better than Fedora or Mandriva or whatever? Of course not. But it's market mindshare is through the roof from a perfect storm of developers that got interested in the things Mark Shuttleworth was preaching. They were NOT tempted by technology. The followed the herd, and lo and behold, now Ubuntu Is doing technically cool things. Now, with all this backing and interest, they ARE moving ahead of Fedora and Mandriva in some core ways. That happens in Free software. It does not ever happen in proprietary software, which is purely driven by corporate interest.

If the ReiserFS falls it will be for the same reason we will eventually have an iPod Killer---because eventually the cool kids that tend to lead the pack will decide there are better things to do. A murder conviction might just cause that. Novell's decision is a symptom of that, not a cause of it.

Tom Caudron
http://tom.digitalelite.com/

Re:Symptoms and Causes

[Anonymous Coward]

> Free Software is about community and community is all about those subjective
> intangibles like reputation, "coolness", and mob effects.

s/Free S/Proprietry s/

> because eventually the cool kids that tend to lead the pack will decide there are better things to do.

WTF? The only "cool kids" buying plastic consumer goods advertised by the likes of Bono are younger teens. Bono and cool just don't sit together but there's another word beginning with 'c' that is quite appropriate.

I've seen posts by you before, you take your narrow world view and provide a high level commentry based on this flawed or limited understanding. Are you trying to come over like an arrogant asshole?

Re:Symptoms and Causes

[borgheron]

Ridiculous.

ReiserFS is an excellent filesystem. It's had it's problems, but which FS hasn't?

Someone will pick up development of it.

GJC

Re:Symptoms and Causes

[anshil]

"Whether we like it or not, this highlights a serious problem with the development model. Likewise, it indirectly highlights one of its strengths. Free software programmers are very much pack animals, like the rest of us. We tend to folllow the herd (I don't mean that in the modern "bad" sense of the phrase. We stick with those we know and enjoy hanging with. We do things we perceive subjectively as fun or cool. We join projects that interest us, we leave projects that offend or dissappoint or bore us. With the GPL, a company that relies on a no-longer-cool project can always pick up the banner and try to reinvigorate interest, but in the end the projects that have momentum have it because they have that special unnamable something that brings people to the fold."

Wheter we like it or not, this highlights a serious problem with the commercial development model. Likeweise, it indirectly highlights one of its strengths. Commercial software programmer are very much pack animals, like the rest of us. We tend to follow the money (I don't mean that in the modern "bad" sense of the phrase. We stick with those we know to pay our loan cheque and enjoy cashing it. We do things we perceive subjectively as uncool, as long the market supports this. We join employers that pay us, we leave employers that offend or dissappoint or dont pay us. With commercial software, a company that relieas on a no-longer-profit-making project can always pick up the huge investions and try to finance it all by themselves, but in the end the projects that have momentum have it because they have that special unnamable something that brings a lot of people to pay for it.

Re:Symptoms and Causes

[anshil]

I've seen posts by you before, you take your narrow world view and provide a high level commentry based on this flawed or limited understanding. Are you trying to come over like an arrogant asshole?

In this case we all humans are arrogant assholes.

Being human you don't have any other choice than taking your world view (things you perceived and experienced) and based on this inevitable limited understanding you project ideas about the higher levels of theories how the world functions, to aid you in future decisions. There is no guarantee this isn't flawed, and luckely we are able of communication to exchange each other about our ideas.. so we can widen our inevitable limted understanding by tapping other understandings.

Re:Symptoms and Causes

[Anonymous Coward]

> Is it better than Fedora or Mandriva or whatever? Of course not.
I'm not trying to start a distro flamewar, but I want to point out that your argument with this example is not really that solid. It is of course arguable that Ubuntu is no better than Fedora, but I wouldn't say it with such a definite tone. Ubuntu does some things better, and other distros do other things better. [ Disclaimer: I'm a die-hard Debian fan ]

> They were NOT tempted by technology. The followed the herd, and lo and behold, now
> Ubuntu Is doing technically cool things.
No. I think what they did was take a technologically solid distro (Debian) and added eye-candy, polish and tweaked the installers and boot screens so that it wouldn't frighten the kids. I don't think Ubuntu has done anything really "techically cool" other than being a "Debian done right". (Above disclaimer applies here)

> Now, with all this backing and interest, they ARE moving ahead of Fedora and Mandriva in some core ways.
Like what? I haven't heard of anything fundamentally new from them.

> It does not ever happen in proprietary software, which is purely driven by corporate interest.
Ask Google, who gets the best engineers in the world partly because the impression that "they are cool".

Re:Symptoms and Causes

[caudron]

WTF? The only "cool kids" buying plastic consumer goods advertised by the likes of Bono are younger teens. Bono and cool just don't sit together but there's another word beginning with 'c' that is quite appropriate.

Not sure I follow your logic here, but I'm sure you understand that the iPod isn't the top of the heap for technical, objective reasons but rather because it's just popular with the right crowd. Either way, the iPod is used in this context as an analogy. Be careful not to overthink anyone's analogies. They tend to break when under too much pressure.

I've seen posts by you before, you take your narrow world view and provide a high level commentry based on this flawed or limited understanding. Are you trying to come over like an arrogant asshole?

I do sometimes come across that way, yes. It's not my intent. Of course, though you ended with a question mark, you weren't really asking a question, but rather just trying to insult me. Not sure what place that has in reasonable discussion. I understand that on the Internet you feel freer to insult people, but it won't win you any real friends or influence the audience to whom you seem to be playing. Rational discourse has no place for the Ad Hominim arguments you are using here.

Sorry you don't like my comments. You can add be as a Foe or whatever and them set an automatic downmod for all foes. I think that'll keep me off your /. radar. I don't set any foes myself, as I tend to like to hear different viewpoints. You may not. That should help solve your problem. Of course, you'll have to log in as a real user to do that. AC's don't get to keep a Friends/Foes list, I don't think.

Tom Caudron
http://tom.digitalelite.com/

Re:Symptoms and Causes

[caudron]

Like what? I haven't heard of anything fundamentally new from them.

When I wrote that I was thinking specifically of the new init stuff they are rolling out with Edgy Eft. Pretty neat stuff, and to the best of my knowledge not found elsewhere. They've done a few things like that, though overall I agree that most of their work is done standing on the shoulders of Debian.

Ask Google, who gets the best engineers in the world partly because the impression that "they are cool".

I stand corrected. That's a good counterexample. I still argue that it's the exception and not the norm, whereas in Free software it's the norm, but my original claim was perhaps too broad a statement. I should remember to avoid "always", "never", and similar boundless terms. They are difficult to defend. :)

Tom Caudron
http://tom.digitalelite.com/

Re:Symptoms and Causes

[dhasenan]

This is TOTALLY bogus. I don't know a single hacker who even took the HURD seriously, or paid it any attention whatsoever.

Stories about Hans began before that

[rbarreira]

And the search for Nina's body, and searches on Hans' house began before that (story published on the 13th of September).

Re:Stories about Hans began before that

[Anonymous Coward]

True, so it is _possible_ this was a straw which broke the camel, so to say (correctly expressed?). The date I read on the missing was 3 septemebre, out of my head though, w/o source. SUSE asked their main ReiserFS (refering to v3) or knew his view, and the change was made. It hits the news only now because of H. Reiser's arrest and the media (that includes /.) loves juicy stories.

Re:Stories about Hans began before that

[Schraegstrichpunkt]

"The straw that broke the camel's back". Also, "september" and "referring".

But...

[rbarreira]

While those events predate the arrest, Hans' house was being searched as early as the 13th of September.

Not a whole of problem

[digitalhermit]

One thing that's great (and not so great) about Linux distributions is that the filesystems are largely decoupled from the block device. Unlike AIX in which the FS and volume type are set just about joined at birth, with Linux you can put almost any filesystem on any device. This means that the tools are more generic (and maybe not as easy to use) but it gives great flexibility. For SuSe, the troubleshooting and command line tools may change slightly, but for the most part there will be no difference for the majority of users. In fact, in my SuSe build I chose ext3 because I'm more comfortable with it.

ReiserFS EOL?

[RAMMS EIN]

Perhaps the reason SUSE will be moving away from ReiserFS has to do not so much with Hans Reiser's arrest, but with the fact that Namesys (developers of ReiserFS) has been focussing their attention and resources on Reiser4 (the successor of the old ReiserFS) for...what? a number of _years_, now?

Re:Murder is not a socially acceptable problem tho

[Anonymous Coward]

I'm glad I flunked you in Philosophy 101.

ReiserFS v3 not maintained by Reiser

[scotsgit]

Novell and previously SUSE have basically been maintaining the version of ReiserFS in the mainline kernel. The choice is easy keep maintaining an FS yourself or use a maintained FS. Whatever the performance or other issues ext3 is the best maintained FS in the kernel.

Ending submissions with an idiotic question

[Gothmolly]

Editors, if you're the ones doing it, please stop. If submitters are doing it, please edit their submissions. We don't need this Roland Piquipaille/Ric Romero style of foolishness, i.e. "Blah blah has happened to company FOO, what do you all think?" Posting it for discussion on Slashdot IMPLIES you're going to get a million different viewpoints, none of which are really important to the submitter. You'll get the viewpoints anyway, you don't need to "prompt" us for them.

Re:Ending submissions with an idiotic question

[Anonymous Coward]

So you also think "Ask Slashdot" has now become obsolete? Puhlease...

Re:Ending submissions with an idiotic question

[volkris]

That has been annoying me for years.

Well, I suppose I have to admit a touch of amusement... I can point at it and say that computer people are dumb.

But overall, annoyance.

Re:Ending submissions with an idiotic question

[75th Trombone]

I tag every single story I see like this "dontaskquestions". Do the same and pass it on.

Re:Ending submissions with an idiotic question

[Mr. Flibble]

I agree with the parent poster, what do you guys think?

Re:Ending submissions with an idiotic question

[Michael Wardle]

Is this the end for our superhero? Tune in tomorrow for the next exciting episode of... Slashdot!

bummer of a downgrade

[louzerr]

Ouch! That would be very sad if SuSE actually backs away from a modern filesystem, to go back to ext2(3,4, etc).

So much for forward progression for Linux.

Maybe we should look at vfat instead ... there used to be a lot of people using it, so it must be good.

A sad day for SuSE, and a sad day for the future of linux.

Re:bummer of a downgrade

[joto]

So exchanging reiserfs for ext3 is a downgrade now? With ext3 you get a stable well-maintained file-system used by the majority of linux users worldwide. With reiserfs you get a poorly maintained file-system, which the original creator has dropped to work on something else, and that nobody else maintains, and that instead of trying to do well as a file-system focuses on performance for other random things, such as being a database, having small files, etc... Maybe you find it sad, I call it common sense!

Re:bummer of a downgrade

[oddfox]

Haha, thanks, I hadn't had a good laugh in a while. ReiserFS is for mail servers, I'd rather have compatability and greater stability (Though neither filesystem has caused me data loss, fsck likes me. Well, except Reiser4, that one didn't) for my desktop though, thanks anyways. I won't even get into technicals like how ReiserFS fragments horribly and quickly. It's a great filesystem that was put into maintenance mode way too soon, they should have done more with it instead of reinventing numerous wheels with Reiser4 and then bitching that it's not making it in the mainline kernel.

Ext3 doesn't seem very slow to me at all, but maybe I've mellowed out with regards to my Linux setups and think most of the time in the past I was simply having a placebo effect feeling. Interesting conclusion drawn here, too:

Conclusion : For quick operations on large file tree, choose Ext3 or XFS. Benchmarks from other authors have supported the use of ReiserFS for operations on large number of small files. However, the present results on a tree comprising thousands of files of various size (10KB to 5MB) suggest than Ext3 or XFS may be more appropriate for real-world file server operations. Even if JFS minimize CPU usage, it should be noted that this FS comes with significantly higher latency for large file tree operations.

Of course since ReiserFS loves CPU cycles, you'd see it perform slightly better when it brings high load to your system during intensive IO.

Re:bummer of a downgrade

[oddfox]

Ugh and I need a cup of coffee! I meant to say you'd see it perform slightly better on a more high-end CPU (than in the linked article) when it brings high load to your system during intensive IO.

Re:bummer of a downgrade

[segedunum]

ReiserFS is for mail servers

Well no. It's ideal for fileserving.

I won't even get into technicals like how ReiserFS fragments horribly and quickly.

Or how much disk space ext3 uses, presumably?

Ext3 doesn't seem very slow to me at all

It is horribly slow, and I've never seen a filesystem create so much disk activity.

Benchmarks from other authors have supported the use of ReiserFS for operations on large number of small files. However, the present results on a tree comprising thousands of files of various size (10KB to 5MB) suggest than Ext3 or XFS may be more appropriate for real-world file server operations.

So every other benchmark that everyone has done has said that ReiserFS is probably the best for thousands of small files, except this one which uses vague terms like 'real-world file server operations'? XFS and JFS would be good fileserving filesystems, especially when you throw very large files into the mix, but ext3? No.

Re:bummer of a downgrade

[oddfox]

Ext3 uses as much disk space as any other traditional filesystem, as far as I know ReiserFS is the exception to the rule as long as you don't mount it with notail since the layout it uses enables it to store many small files in a single block. Tail packing increases external fragmentation and introduces a speed hit for an, on average, 5% increase in storage copacity when compared with ext2 (And I would presume ext3 since ext3 is simply a journal on top). ReiserFS may be good at fileserving, but the fact of the matter is it's best when dealing with tons of small files a lot of the time for maximum throughput, mailservers. I have never seen anyone argue that ReiserFS is not ideal for mailservers. Ext3 mounted in writeback mode is dangerous for poorly written programs but delivers solid performance in a lot of scenarios. Guru Labs tests

In any case, for what it's worth I use ext3 in journaled mode and it performs quite fast enough for my needs and I don't see "so much disk activity" but then again I'm using the system as a desktop/wanna-server, I'm not running benchmarks on it all the time simulating rediculous scenarios. I don't trust XFS because of how it aggressively makes use of RAM and a poorly designed program runs a good chance of losing data in the event of a power outage, and JFS is great for saving CPU cycles but it's throughput and latency isn't the best, though it's probably my second favorite journalled FS after ext3. Ext3 is reliable for me and when using dir_index and journalled data mode it does alright by me in terms of performance, and I enjoy the security of having full data and metadata journalling, which AFAIK ext3 is the only one that provides.

Re:bummer of a downgrade

[industrialvegan]

Agreed. It is another sad milestone on the road to annilhilation of my favorite distro. Not to mention the outlook for Hans. That seems perilous indeed.

I was not happy when Novell removed the ability to choose JFS during installation. I've got 5 servers and two workstations using it(running 9.2) an not once have I ever had a problem. In all fairness, I haven't had any problems with ReiserFS either, but downgrading to ext(x) does seem like one more nail in the coffin of a once mighty distro. Hell, even Ubuntu offers JFS as a choice during installation. Although, Reiser was the default, if I recall. So what could be so bad about JFS that Novell decided to stop supporting it when other distros still offer it and there doesn't seem to be any compelling reason to abandon it? The only thing my little mind has been able to come up with is the IBM connection and that that did not somehow sit well with Novell. The only other conspiracy I can come up with is that Novell is mining Open Source software for everything they can and in the process are slowly strangling SuSE to the point that it will finally collapse and be abandoned in favor of something more tightly integrated with Novell intellectual property.

uncertainty

[glas_gow]

I don't think it's so much that ext3 is more stable than Reiserfs, just that extX has a future (or two futures, to be more precise) whereas the future of ReiserX is a little uncertain at the moment.

Novell?

[Anonymous Coward]

Who's Novell and what do they have to do with Linux?

Re:Novell?

[funwithBSD]

They owned UNIXWARE.

Too Silly! Too silly! Stop it this instance!

ext3 to match ReiserFS 4?

[GrievousMistake]

That does seem a little unlikely. Isn't ext3 still basically ext2 with journaling? How are they still making such progress with it that performance will soon match a modern filesystem like Reiser4, which among other things has a more optimized disk layout and will have transparent compression? If there are patches to bring those to ext3, they're neither stable nor ext2 compatible, which are supposed to be ext3's advantage.
(Meh. Upon RTFA'ing I see what they meant was that ext3 will "soon" match the performance of ReiserFS (3), and that it is still more stable than Reiser4. The summary still deserves the rant, and I'm actually curious about how they are improving performance in ext3 nowadays, so I'm still posting this.)

Re:ext3 to match ReiserFS 4?

[quinnharris]

Hans vision (http://www.namesys.com/whitepaper.html) is about unifying namespaces (fs, database, email, config files, etc.). So I believe his primary objective was to build a file system that performs well under all situations especially those situations where current filesystems perform so bad nobody writes software to use the fs that way. Most notably lots of tiny files.
Ext(2/3) use a traditional UNIX FS design and most software has been written to work acceptably on this type of file systems. In other words, nobody writes software that creates thousands to millions of tiny files because ext absolutely sucks at this (and ext4 doesn't help much) . So to get around this most of today's software create a new namespace in a single file. /etc is exactly this.
reiser(fs/4) also use extents which improves its handling of really large files which is becoming more common (movies, music, etc.). ext4 introduces this feature.
So for today's common usage patterns and file sizes ext4 will perform reasonably well compared to reiser4.

But if you where to extend fs semantics enough to make it reasonable to use the fs to directly back something like gconf (http://en.wikipedia.org/wiki/GConf), reiser4 would be substantially (10x) faster than ext(2,3,4) could ever hope to be. In principal if you extend the filesystem to directly support gconf and similar you would be able use of many fs tools for free.

Consider the windows registry. It is a mistake because it obscures the information by its convoluted structure and just as significantly, one can't use all those usefull tools on the registry you use with your file system (well not to many in Windows). Gconf addresses the first problem but not the second. Overall, I would consider the /etc approach superior because it is better documented, more transparent and I can use cp and subversion on it. Nobody will ever write a tool as powerful as subversion for the windows registry because the value isn't great enough. But subversion was never designed specifically to help with /etc it was primarily for coding but because coding has much in common with /etc the same tool can be used for both.

Yet, the windows registry had the good idea of creating a consistent way to store the the same type of data. Saving programming time and building a single consistent source for all things configuration. I believe it is possible to build something that gets the best of /etc and the best of the windows registry or better yet gconf.

The more use one can get out of a tool, the more people will work to develop and improve it. By using the same namespace for as much as possible, the more value a tool designed to work with that namespace will have. In other words, the utility of a computer system is proportional to the number of ways the components can interact, not the number of components.

To achieve this objective, one needs a solid foundation, a system that can efficiently store whatever you ask of it, that is what reiser4 is trying to be. But, clearly much work remains.

incorrect

[Anonymous Coward]

And Reiser? IIRC (correct me if I'm wrong please) its even longer around than xfs

Hmm. You are incorrect except in one sense. I have been using XFS on SGI systems for 10 years, so it's older. However, it may have been introduced into Linux later then ReiserFS.

Re:Murder is not a socially acceptable problem tho

[Anonymous Coward]

Yes, and thanks to that, i took a job as head of education in your area, expect a pink slip tomorrow....

That's fine.

[Neuropol]

Good. I'm about to move away from Linux entireley, myself.

I'm tired of beating my head against the wall and jump through hoops to get things working.

I've never wasted so much time in my life.

What a let down.

Re:That's fine.

[Ant P.]

That's funny... exactly the same reasons I dumped Windows.

Re:That's fine.

[Frumious Wombat]

He didn't say he was going to Windows. He could be a Frustrated AIX user. There's also the VM/CMS community which has trouble adapting to Linux as well. Something about not having a card punch and six IBM tech's in the box when they install it.

Re:That's fine.

[cool_arrow]

Stop crying and switch. Troll.

Re:That's fine.

[TheZorch]

Other than the fact that a lot of control setting screens are different and are in different locations I've found usng Linux to be easy. Its a myth perpetuated by Microsoft that Linux is hard to use. Setting up Linux and configuring it hasn't been more easier in the past few years than setting up and configuring Windows. Some of the configuration tools are different, some users still don't understand the concept of partitions, and the majority of the reasons why a lot of people are afraid to switch is because of all of te FUD that Microsoft has spread either themselves or through computer industry reporters they bought out.

One need only look to the financial ties between Microsoft and SCO to see the truth.

Re:That's fine.

[Anonymous Coward]

the majority of the reasons why a lot of people are afraid to switch is because of all of te FUD that Microsoft has spread either themselves or through computer industry reporters they bought out.

That's great; when someone has a problem with windows slashdot fags just assume "sure, everyone does" but when someone has problems with linux you all bitch about it being FUD or being bought. Bullshit. Simply bullshit. WTF, you're just another linux fag. It's people like you who alienate the OSS community from the masses.

Re:Slackware

[Anonymous Coward]

It would also be foolish to choose Slackware as the GNU/Linux distribution for a large organisation.

No one laughed last time...

[Chapter80]

I wonder if Reiser 4 "file" system is hidden inside of a cake.

Re:No one laughed last time...

[Anonymous Coward]

I hate to say it, but I don't get it.

Posting as AC, because believe me, I feel bad that I didn't understand a +5 funny :/

Re:No one laughed last time...

[jadavis]

The cliche way to break out of prison is to have someone bake you a cake with a file (a metal file, not a nuch of bytes) in it. The cake hides the file. Then, after you eat the cake, you use the file to break out of prison. This probably would not actually work, but I'm sure it's happened before.

Re:No one laughed last time...

[chris_sawtell]

not a cake, a laptop surely?

Re:Murder is not a socially acceptable problem tho

[AusIV]

Juries don't prove people guilty, prosecutors do. Juries find people guilty.

political or technical motivation?

[Anonymous Coward]

ext3 matching reiserfs performance soon? Hmm... One may only wonder how much political motivation there was behind Novell's decision.

Don't be a hater man!

[Anonymous Coward]

Some bitches needs a cappin'!

disagree

[diegocgteleline.es]

SUSE decided to change to ext3 BEFORE hans was arrested, and did it for technical reasons

Re:disagree

[Anonymous Coward]

They want to run out of inodes and force a lock-in path to force people to upgrade?

didn't work out for me...

[shikhs]

i was using reiser since it was default with suse 10.1. i moved to ubuntu, for which i had do to a hard-reboot once - and it caused errors fsck just couldn't repair and i couldn't boot. i could access files from a windows reiserfs tool however, so recovered some and i'm happy with ext3 and ubuntu :)

We'll match ReiserFS "Real Soon Now"

[borgheron]

Yeah, sure... call me when that happens.

GJC

controversy

[toby]

Due to the controversy that always surrounds anything I post, I never bother reading comments for my posts.

Oh, you have that problem too! You must be an Independent Thinker -- closest thing to a terrorist!

Actually...

[absurdist]

...your comment would be more correct if you dropped the "For a white guy?" part...

This has nothing to do with the arrest

[Anonymous Coward]

Dumbass

From the Horse's Mouth

[Jeff Mahoney]

I wrote the original email proposing that SUSE switch from reiserfs to ext3. At the risk of triggering responses of "The lady doth protest too much," I'll restate a few statements I've made elsewhere in response to common questions:

1) The decision has *nothing* to do with Hans' situation. The email was released on the same day as the initial story broke, but it was pointed out to me after I had sent the email. I was concerned then, correctly as it turns out, that people would consider the two issues intertwined. They're not. My proposal was based on technical and maintainability reasons alone. The timing is an extremely unfortunate coincidence.

2) SUSE is *not* dropping reiser3 support. This change only affects the default. It doesn't change our support of reiser3 at all. We still support four major file systems: ext3, reiserfs, xfs, and ocfs2. Our installer offers other file systems as well as a convenience, and users are free to use any of them. So, if you're committed to reiser3 or xfs, nothing is stopping you from continuing to deploy systems using them.

3) Many benchmarks show reiser3 as performing better than ext3, and this is mostly true. What isn't shown in those benchmarks is that if you're operating two or more reiser3 file systems in parallel, performance will degrade for both of them due to the use of BKL everywhere. ext3 (and other file systems) will don't degrade in that case. I've also read reports that there is a bit of research going on into making ext3 locking finer grained. I don't have any sources to cite, but any reduction of critical sections without reducing reliability is always a good thing.

People refer to reiser3 as a modern file system, but I'd call it progressive. Reiser3 has served us well for years, but it's showing its age. The basic idea behind reiser3 is still sound, and when extended with integrated integrity checking and better b-tree locking borrowed from years of database research, it would perform extremely well. The problem is that adding the first is a huge disk format change, which means it's no longer reiser3. Adding the second is a hugely invasive change that would throw out a good chunk of the existing code -- again, essentially creating a new file system. It would be like people saying, "I like my ext3 file system, but I don't like the code. Let's start over." Combined with a small development community, it's a recipe for instability and there are more interesting problems out there.

I've posted some more lengthy comments here: http://linux.wordpress.com/2006/09/27/suse-102-dit ching-reiserfs-as-it-default-fs/#comment-28534

What's ReiserFS?

[filesiteguy]

Honestly, I've been using SUSE for a few years now as my full-time system. I couldn't tell you that I was using reiser except for the fact that I read up on this.

In other words, for most of us everyday SUSE users, I doubt a switch from Reiser to EXT or JFS or whatever will mean anything. As long as my system works and works faster than Wintendo, I'm happy.

Oh, and for the record - I think OJ did it.

Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[mosel-saar-ruwer]

Did Novell ever get around to porting Novell Storage Services [NSS] to Linux?

NSS was the B-Tree successor to the old allocation table NetWare file system, and it had all the permissions and attributes that were unique to the Novell World:

Read
Write
Create
Erase
Modify
File Scan
Access Control
Supervisor

So did Novell ever get around to porting an R/W/C/E/M/FS/AC/S file system to Linux, to be used in place of the standard Unix RWX/RWX/RWX file system?

And if so, is anyone out there using it?

Re:Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[certain death]

Well...I am willing to bet that the options are there is you are using one of their fantastic GUIs, say maybe NWADMIN, but do they actually work?, Did they EVER actually work?...Sorry, I guess that is more of a question than an answer to your question.

Re:Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[T-Ranger]

Yes, and yes. It comes with Open Enterprise Server and remains closed source. Its primary usage would continue to be, Id wager, in a 1:1: replacement servers for old Netware systems; e.g. desktop office support.

Its interesting that a lot of "high-level" CMS's all implement their own ACL system, but for users/groups as well as content (files). For all the interesting and directly-usefull-to-users apps Novell is building (e.g. Beagle) I've been thinking that it would do them good to build some libraries and proof-of-concept apps that leverage (sory) their excelent low level capabilities. Its just that most developers dont know any better..

Re:Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[SysPig]

Did Novell ever get around to porting Novell Storage Services [NSS] to Linux?

Yup...it happened with their initial release of Open Enterprise Server, well over a year ago. OES is built upon SLES9.

That being said, we found the initial releases to be a bit buggy, and performance was way off what we've come to expect with Netware. However, things have improved considerably since the initial release, and we're running four OES 1.0/SP2 Linux servers with NSS now for sites with moderate loads (file/print for up to 100 users.) Benchmarks show things are still a bit slower than Netware, but not enough that the users are complaining.

The other issue with NSS on OES Linux - backup support. Relatively few vendors have supported backup solutions, compared to "standard" Linux filesystems. Standard backup tools handle the file data just fine, but they won't capture the NSS extended attributes - trustees, ownership, etc. There are backup solutions that work well, but they aren't free.

Good news is, it's allowing us ot migrate off of Netware, without giving up the Novell services that work very well for us. It's also transparent to the users - with NSS on OES Linux, the servers look and behave just like a Netware server to client PC's.

Re:Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[Foolhardy]

Its interesting that a lot of "high-level" CMS's all implement their own ACL system, but for users/groups as well as content (files). For all the interesting and directly-usefull-to-users apps Novell is building (e.g. Beagle) I've been thinking that it would do them good to build some libraries and proof-of-concept apps that leverage (sory) their excelent low level capabilities. Its just that most developers dont know any better..

I'm afraid that all the major operating systems have many developers guilty of using the OS's low level facilites either incorrectly or reinventing them, especially when it comes to access control.

Windows NT's security model is quite powerful and reusable but almost all devs (Microsoft's own included) misuse or ignore it, and we all know how well that's working out...

Re:Novell Storage Services [NSS] R/WC/E/M/FS/AC/S?

[T-Ranger]

True, true.

The reason in Microsofts case being that the model is extreemly complex, and historicaly (anyway) poorly documented. Sure, some people dont use it because they dont know about it, but a a lot of people dont use it because its just too fucking difficult to learn, and get right.

ext3? sucks!!!

[Anonymous Coward]

In all tests I've done ext3 has shown it self _the slowest_ filesystem of all (xfs, jfs, reiserfs), plus god forbid if you have like a few terabyte partition to have it under ext3! Then you'll wait FOR DAYS in case it decides to do a full fsck! I was just starting moving away from Mandriva and to SuSe, and this is really bad news to me, since looks like the move will have to wait until I see whether or not it'll be like in RedHat (totally no reiserfs during the installation), or it'll be an option. If it's like RedHat - absolutely no, dead deal. No Suse on my servers. Plus, they suck anyway, when compared to Mandriva: VERY small choice of packages supplied, plus it's impossible to rebuild a lot of packages from the sources they provide due to unsatisfied dependencies: some packages are missing, some are wrong version...

MPU

[Ignignot]

I'm tired of this crap. I have a feeling that editors just feel that they have to add something to submissions, so they add in "what do yall think???" at the end.

RTFM

[mikaelhg]

Its the same deal with Sun's current development ZFS: it lacks the option to decently make a backup.

See Solaris ZFS Administration Guide, Chapter 6 Working With ZFS Snapshots and Clones.

Wasted space

[kimvette]

ReiserFS is touted to be a zero-slack filesystem, whereas ext3 still ties up entire blocks (groups of sectors, usually 4K, 8K, 16K depending on formatting options) on, say, a 1k text file, or a file which spills over into a tiny fraction of another block. When you have thousands of files which take up only a portion of a block, resulted in a lot of wasted space (how many files are exact multiples of the block size?). Some may argue "yeah, but disk space is cheap" but even so, 750GB drives are the largest we can buy now (yeah I know, RAIDs, but the point still remains), and if you fill it up doing projects and need just 80MB additional space to complete a job, a 300MB of allocated space is unused portions of blocks (slack), it sucks knowing that you could have finished a job if the filesystem weren't so inefficient.

Re:Wasted space

[Anonymous Coward]

As previous posts have already mentioned, reiser3 uses the BKL rather intensively. Now consider that it _also_ has to do the math for file packing.

Strange Original Decision for Novell

[Smackintosh]

When I was introduced to the new Novell releases based around Linux I was very surprised indeed that ReiserFS was the default. I know you can't judge a book (in this case Hans and his filesystem) by its cover, but I think a public company would make more conservative decisions given some of the negativity (both technical and non-technical) related to ReiserFS.

I always had the impression that the filesystem and its development were more cutting edge than a commericial software company would feel comfortable with.

Re:Strange Original Decision for Novell

[Wudbaer]

reiserfs has been the default filesystem on SuSE for years, so it was not a Novell decision. They just kept what had been working for the original SuSE.

Re:Strange Original Decision for Novell

[Smackintosh]

Well that completely makes sense. Thanks for the clarification.

So I'm guessing the decision to move from ReiserFS to ext3 is then the expected conservative one by a commercial software company.

It just begs to be said...

[supabeast!]

What implications will this have for SuSE users...

One-third of them will have to reformat and reinstall to get everything on ext3. The other two SuSE users won't have to change anything.

bada-BING!

LOL

[Anonymous Coward]

Linux is STILL for fags.

Rubbish

[segedunum]

Ext3, in contrast, is stable and likely will match ReiserFS's performance advantages "soon."

Rubbish. Ext3 has never been able to match Reiser's performance on small files or in other areas, and the notion that ext3 is going to match it is absurd. Even ext4 is not likely to catch up. A lot of ext developers have bizarre ideas about how their filesystem compares to Reiser, XFS or even JFS in a lot of areas. Ext is simply a stable and solid, but badly evolved, filesystem and it is a filesystem that generates an awful lot of disk activity.

Re:Murder is not a socially acceptable problem tho

[Cylix]

Maybe where you live that's true...

Not just the name...

[LinuxGeek]

I've been to the namesys website and haven't found anything about the Hans Reiser arrest. If it is more than a one man show, then they should have a prominent statement about their intention to continue development regardless of the outcome. Not seeing something to that effect after this mych time would make me quite nervous if I had a business or product that relied on continued development of the ReiserFS line. Seems as if Namesys is accepting the inevitable demise of the whole organization at this point; I hope that changes.

XFS failure

[Sloppy]

This means that in case of a power cut, XFS may leave the contents of a "file" (metadata) unscycrhonized with its data. Because of that, the metadata may be pointing to random free zone of the disc with confidential information (passwords) which was deleted but it has not been overwritten, so XFS sets it to zero for safety.

Does that mean if I mkfs.xfs with -d unwritten=0 I might potentially avoid the corruption problem in exchange for taking the security hit? Or does it mean I would still likely (assuming that failure scenario happens) have corruption, but with the symptom being files containing wrong contents instead of zeros?

XFS

[Xaero_Vincent]

I havent experienced any data loss with XFS when I do an improper shutdown or reboot. This is probably because I disabled write-caching. So now changes are immediately written to the disk rather than the disk cache, but at a cost of performance. In the event of a power outage or crash data would already be permanently stored.

ReiserFS is great, don't kill it, just rename it!

[fromvap]

The good thing about open source is that if a person goes rogue, their code can be taken over by someone else and their existence erased from history. Just call it "NewFS" or something. "Someday we may catch up with ReiserFS" That is insane. Just use ReiserFS and give it a new name.

Specialists vs generalists

[Sloppy]

it claims ext3 .. will 'soon' match performance with the newer ReiserFS 4.

This simply won't happen. There are lots of choices in filesystem development, and if your application doesn't match the choices that were made, then that filesystem won't be best, or "match" the specialist that did make matching choices.

There is no way that Extn will ever match, for example, ReiserFS' performance on working with a directory full of ten thousand 700-byte-long files. ReiserFS will do directory-related things faster, and tail-compression will save you space (and therefore give you even more performance, thanks to caching).

I don't have a problem with SUSE picking something else, though, because my whole point is that, no matter what FS you pick, if the default configuration is that the installer just formats the whole disk as one filesystem, then no filesystem is going to be ideal in all cases.

Of course, the Gentoo Ricer approach is to break your disk array up into little pieces, so you're using performance-over-safety filesystems on the RAID0 parts, using safety-over-performance filesystems on the RAID1 and RAID10 parts, and compromise filesystems on the RAID5 parts -- and within each group there is a variety of different formatting and mounting options used. (Not to mention a little tmpfs here and there; not everything has to survive a reboot.) Yeah, df lists 20 different mountpoints, every part of the hierarchy "optimized" (*cough*) for what it gets used for.

Now I just need some good-looking stickers to put on the outside of the case, and it'll be even faster! Yeah, next weekend I'll probably spend a few more hours changing something, but for the next 5 days I'll be pretty smug about every millisecond I save.

UPS doesn't protect from kernel panics

[myowntrueself]

A UPS doesn't protect you from kernel panics.

Having ext3 mounted 'data=journal' for your /var/log filesystem can let you see actual log entries from the time leading up to the panic, as opposed to log files with a block of crap in them.

And if anyone doubts the performance of 'data=journal' run iozone to do some comparisons; if you are making so many writes so quickly to large enough files for it to make a noticable performance difference then you already have big problems with your logging.

Default FS choice.

[itomato]

Unless you're using a Novell Linux product (SLES/SLED 8, 9, or 10) which is what this article (Novell Moves Away From ReiserFS) is about.

The default in those cases is ReiserFS.

I know of a project that implemented that same logic during the planning and initial implementation phase.

They will now be exploring the possibility of a filesystem change.

Re:Murder is not a socially acceptable problem tho

[Overly Critical Guy]

It's kind of hard not to consider him guilty when the police found splatters of Nina's blood in his house.

Wait a minute...

[soccerisgod]

I did read about that before that murder story. I think it was on Heise.de. I think that conclusively proves it has nothing to do with that.

Ah yes, here we go: http://www.heise.de/newsticker/result.xhtml?url=%2 Fnewsticker%2Fmeldung%2F79035&words=Reiser - that was dated 10/04, 7 days before the murder story broke.

Re:Wait a minute...

[Cherita Chen]

You are absolutly right. The decision to move to ext3 as the default FS was made quite some time ago. One can't help but laugh at all the "informed" /.'ers

How about a nice cup of "Shut the fuck up"?

[SaDan]

Be quiet, you EXT2/3-bot. EXT2 and 3 are not bullet proof, neither is ReiserFS. If there was one "perfect" filesystem, I believe we'd all be running it by now everywhere.

Take some of your own advice, and stick with your distro default. My distro defaults to ReiserFS.

Re:How about a nice cup of "Shut the fuck up"?

[Anonymous Coward]

EXT2 and 3 are not bullet proof

Having trouble reading are we, zealot fuckwit. He wrote: "virtually bullet proof". The Ext2/3 record shows this cleary -- and only an idiot would disagree with it. Unlike Resier, which has a documented history of poor recovery tools, poor support, brittle handling of hardware problems and massive support costs (reformatting etc). You'd think that Novell's switch and the technical statement in this article would actually penetrate the thick skulls of Reiser-bots... but no. It just makes them louder and more faithful.

Re:How about a nice cup of "Shut the fuck up"?

[SaDan]

I've had plenty of trouble with EXT2 and 3 on hardware that performed just fine with ReiserFS. I can say the same the other direction too. I prefer using ReiserFS, I've had very few issues recovering failed arrays and drives using ReiserFS. I've had more issues recovering EXT2/3 filesystems in what should have been less severe failures. Performance for the type of work the machines I support do is MUCH better under ReiserFS than EXT3, even after all the speed tweaks I could throw at EXT3.

Go fuck yourself and the filesystem you rode in on. You've added nothing to discussion, anonymous dimwit.

Re:How about a nice cup of "Shut the fuck up"?

[Anonymous Coward]

Well, in terms of linux I personally prefer EXT3. It recovers fairly well from a power off which is more than I can say about EXT2. When I first tried linux, I remember losing half my file sytem to a power loss with EXT2. To this day, when someone brings up linux I think back to that moment when Windows NT suddely became more reliable in my mind. Before anyone makes a comment about a UPS, this was 1998 and I was working for an ISP making 7 dollars an hour and living with my mother in a trailerpark. It was a miracle I even had two PCs.

At least there are several choices for journalized file systems in linux. In the BSD world, all we have is soft updates and a buggy gjournal implementation. Be thankful for what you have and work hard to improve it.

BKL == Big Kernel Lock?

[RAMMS EIN]

``What isn't shown in those benchmarks is that if you're operating two or more reiser3 file systems in parallel, performance will degrade for both of them due to the use of BKL everywhere.''

Where BKL is Big Kernel Lock? It wouldn't be the Big Kernel Lock that's the culprit, again. Is that also what causes the CPU to be in wa (as shown by top) a large percentage of the time when reading from slow media?

Re:BKL == Big Kernel Lock?

[Jeff Mahoney]

Yes, BKL is Big Kernel Lock. If you look at the code, you'll see reiserfs_write_lock(sb) everywhere. This a macro for lock_kernel(), which is the Big Kernel Lock. It's a regular spinlock[1] with two interesting side effects. It's dropped when the thread sleeps and is reacquired when the process reawakens, and you're allowed recursive locks[2]. Since the kernel will schedule when it's waiting on a slow disk, it drops the lock, so that's not a source of contention. The real issue is when it's doing things like balancing the tree or other CPU-intensive task, it's holding the BKL - and that means completely unrelated file systems can't do anything.

To get back to your other question: When you're looking at your top(1) output, since a spinlock is just a tight loop waiting for a value to change, it would appear as sy. When the kernel is waiting on a slow disk, it appears as wa.

1: There is a config option that uses a semaphore instead of a spin lock, but I'm not sure how widely used it is. It would appear as sleeping rather than system execution.
2: This is the hard part about simply replacing the BKL with another fs-specific lock. Other lock types aren't recursive.

Re:BKL == Big Kernel Lock?

[RAMMS EIN]

Thanks for the enlightenment!

Correction

[Anonymous Coward]

"Novell and ReiserFS - Once you're with Reiser, you'll never want to leave."

How's this:

"Novell and ReiserFS - Once you're with Reiser you'll never leave."

Re:Correction

[tomhudson]

Makes it sound like the new "Hotel California".

Q: What happens when you cross SubVersion with ReiserFS?
A: You can check your files out any time you like, but they'll never leave ...

Please, Linus, Don't!

[Nicopa]

Please, Linus, don't murder anybody!

                                                  -- a Linux user.

Re:Please, Linus, Don't!

[trupoet]

ROFL thats great.

Took me a minute to get it.

You were not right, this is no kneejerk reaction

[Chuck Chunder]

This particular change has been in the pipeline for over a month

ZFS

[Anonymous Coward]

When are we going to see ZFS on Linux? It pwns all these filesystems put together.

I can relate

[Trogre]

Late last year I was researching Reiserfs and Ext3 to see which would be best suited for my new server.

Resierfs looked like the clear winner for two good reasons:

1. Reiserfs is faster. Much faster than ext3 in nearly every scenario. Large files and small files.
2. No inode problems. If your users fill your HD with hundreds of thousands of tiny files you're not going to run out of inodes before you run out of disk space. This is something that needs to be anticipated (at the cost of more disk space) at filesystem creation time in ext3.

Reliability for both filesystems was pretty much the same from all accounts.

But in the end I went with ext3 for one and only one reason: Recoverability.

Reiserfs had no, or very few decent, recovery utilities. If a filesystem corruption occurred (and it seemed that the probablity of such corruptions was equal for both filesystems), then data on an ext3 fs stands a much better chance of being recovered than on a reiserfs one.

Of course that was late 2005; that situation may have changed by now.

Re:I can relate

[Anonymous Coward]

IIRC the problem with reiser3 is that it has only one root node (as in binary tree, not directory). Tamper with that and go hoping that fsck can find everything again. Other fs such as xfs (maybe ext3 and others too, no idea) that create allocation groups have it split.

Re:I can relate

[SanityInAnarchy]

No inode problems.

Common misconception. If anything, Reiser3 has more problems with this than ext3, they just aren't called "inodes".

The difference is, you can run out of whatever-they're-called -- and it is a truly technical distinction -- without ANY advance warning. At least with ext3, you can plan for it, and add more inodes if needed.

I've actually had VERY good luck recovering data from a Reiser4 filesystem. There's even a decent undelete feature, enabling me to recover from the equivalent of "rm -rf /" (don't ask), so that's not such a huge deal for me. But as far as the inode issue goes, you'll be much better off with Reiser4 -- as far as I know, it actually is designed the way you suggest -- it will run out of disk space before it runs out of potential files ("inodes" or whatever it calls them).

so that means...

[Anonymous Coward]

He must have killed the bitch if you randomly got nulled out files in your home directory...he must pay for said crimes upon humanity.

Heres one good reason why not (XFS)

[myowntrueself]

why not move to xfs?

Heres one single good reason.

XFS is very reliable, but not 100% reliable. Sometime you may have to run xfs_repair

If you have an XFS filesystem of more than a couple of hundred gigs you will need more than 4G of RAM to run xfs_repair. The cut-off point is not well defined so you won't know how big was too big until its too late.

The XFS people will tell you to take your disks and put them into a machine with enough RAM and do the fsck there. Believe me, I have seen XFS developers actually hand this out as instructions on how to repair an XFS filesystem. I thought it was a joke.

This is an unworkable solution for oh so many reasons, but heres the killer.

Suppose that you have a RAID array, running on hardware RAID which is on the motherboard and your XFS filesystem is on this array. The hardware is not 64 bit nor PAE and you don't have enough RAM to repair the filesystem.

So you take your disks and install them into a 64 bit machine with plenty of RAM and oh wait it was hardware RAID and the controller on this board is different, oops. Time to find out if your backups work.

I love XFS but there is such a thing as too much of a good thing :)

I beg to differ.

[CrimsonScythe]

Any good file system has proper lookahead. This lookahead made Novell capable of knowing that Hans Reiser would indeed be arrested. Had they used Reiser4, they could probably have been able to look even further ahead and predict that murder several months ago. In comparison, my box running Ext2 just predicted that we'll have a second president from the Bush family. Now, that's just silly, so I guess it's about time to upgrade to Reiser4...

And you are still wrong

[jotaeleemeese]

As explained by other people on this thread, that decision was taken some time ago on technical merits.

The name of a fucking filesystem tainted, yeah, sure.

If 1% of people with some IT expertise know what reisrfs is, I would be surprised.

How do you taint something nobody knows about?

And the people that know about reiserfs I think can be trusted to take technical decisions on their own merits.

Or do you think MS is going to launch a campaign saying "Linux is made by criminals"? (if, and that is a huge if, Mr Reiser is convicted).

Re:And you are still wrong

[Anonymous Coward]

As explained by other people on this thread, that decision was taken some time ago on technical merits.

And, as explained by other people, after the disappearance of Nina Reiser and suspicions falling on Hans Reiser. In fact, the timing, one day after the police searched his home, looks suspiciously like a knee-jerk reaction.

If 1% of people with some IT expertise know what reisrfs is, I would be surprised.

How do you taint something nobody knows about?

And the people that know about reiserfs I think can be trusted to take technical decisions on their own merits.

You think wrong, or not enough, as may be. Most decisions are taken on a level where the decision maker does not know anything about file systems at all, but they do listen when a salesman tells them "You want to go with SuSE? You can't trust that. Even their file system is made by a, and I don't say this lightly, murderer! Yes, I can document this. There's no felons working on our products!"

The suits who make the decisions don't have to know of any taint; they will be told.
It's just common business sense for vendors to want to distance itself, no matter whether there's a chance he's innocent or not, or whether they cite the real reason or not. In most cases, I imagine they won't, or they'll be seen as insensitive by one camp and too late by the opposite camp.

--
*Art

This opens the door for SELinux

[plazman30]

Doesn't SELinux require the use of ext3 or xfs?

Just to finish your reputation as a futorologist

[jotaeleemeese]

http://linux.slashdot.org/comments.pl?sid=200813&c id=16444091

modder test

[nortcele]

This was modded flamebait.

Silly modders. Let me show you flamebait.

Reiserfs has a unique algorithm that can find the 'kill' file quicker than xfs, ext2, or ext3.