Data Loss Bug In OS X 10.5 Leopard

An anonymous reader writes "Leopard's Finder has a glaring bug in its directory-moving code, leading to horrendous data loss if a destination volume disappears while a move operation is in progress. This author first came across it when Samba crashed while he was moving a directory from his desktop over to a Samba mount on his FreeBSD server."

Comments

That's silly.

[ackthpt]

Normally while moving you ensure the copy completed before deleting the original. Apple must be using some discount programmers.

Re:That's silly.

[Hatta]

Why don't you just use rsync?

Re:That's silly.

[nmb3000]

Apple must be using some discount programmers

Of course not! Don't be a troll.

Everyone knows that Apple's products Just Work, and that's no different in this case. The files were moved just like you asked, and if you can't find them. well, that's not Apple's fault, is it? You don't blame the contractor who built you home when you lose your keys, do you?

In any case, you should be using Shadow Copy...er...Time Machine which would have protected you from going and losing track of your own files.

Re:That's silly.

[Eq 7-2521]

No, it's probably just that the automated Schnapps IV system that they have for maintaining the Ballmer Peak went awry one day. (Read the tooltip if you haven't seen this before.)

Re:That's silly.

[Frothy Walrus]

Oh yeah, rsync. Is that one still broken on 10.5? Apple's build of rsync on 10.4 consistently choked on a few files when my dad started using it on his Mac Pro.

Re:That's silly.

[doxology]

then some metadata on the disk is modified to reflect your move.

Re:That's silly.

[ackthpt]

So what happens if you're moving a 120GB folder one directory level on a 150GB disk?

Typically if you are moving within the same logical device the file pointer is moved and no copying need take place.

When moving to another device your code reads and writes, within a loop and traps exceptions (such as the device suddenly vanished, where the OS should raise an exception and your application traps it.) A wide variety of errors could occur while moving and in the event any of them happen the user should be notified in an appropriate manner and the original data not deleted.

I've written a number of applications which moved files or data between databases and it's fundamental your application is on the watch for any problems. Not to have an exception raised or to trap any and all, well, that's simply an inexcusable lapse.

This sort of thing is extremely critical if you happend to be defragmenting a disk drive. Long before Macs and PCs we had to defrag our mainframe drives and the applications which did the work were quite careful. Often the best practice, if you had the resource of a second drive, was to simply defrag to a new drive then re-assign the new dist as the original.

Re:That's silly.

[Taagehornet]

this is not what current generation of typical user would do, but I believe they should be educated on this anyway

Reeducate the user, you say. Surely you must be joking, right?

Let's ignore for a moment that Leopard may have a few bugs that will have to be ironed out. That's only to be expected with *_any_* newly released OS and the reason why no sane person would ever dare to update the OS on a mission critical machine within the first few months of the release.

However, if you can't rely on your OS to perform a simple file move without risking data corruption, then the right solution is definitely not to verify every single operation by hand. Automating tedious tasks is exactly what computers do best, and that the OS ensures the integrity of the copy before throwing away the original is definitely something you should expect.

Re:That's silly.

[Matt Perry]

I think you misunderstand something.

No, I understand perfectly.

I know for a fact that I've successfully canceled FTP-moves on Windows (brand); I also know for a fact that I've successfully canceled FTP-moves on all modern instances of the Windows OS with various FTP clients (programs).

I'm not talking about other clients. I'm talking about using Windows XP as the client.

What program were you using that fragged your FTP server directory exactly?

Windows XP, service pack 1.

I suspect you're a fud-fucking idiot

You need to work on your social skills.

although at the moment, I'm interested to hear if there's more than fabrication here. Pray tell: how can I reproduce this bug?

You might need to use WinXP service pack 1. I'm unable to reproduce it now on my home machine which has SP2 installed. At work we've just upgraded to service pack 2 within the last six months. Anyway, about a year ago I was trying to move nearly 40 gigs of data from a Solaris server at another site to my local machine. The Solaris server was an FTP server and that's the only access I had to it. I opened Internet Explorer and typed in ftp://servername and was prompted for a username and password which I entered. Then I saw the files in the explorer view just like I see files on my local machine. Since I was moving so much data I decided to do it one top-level directory at a time. To move the files I selected the folder, dragged it to a folder on my local machine, and held down shift when I let go of the mouse pointer so it would move instead of copy. I noticed that Windows went through three steps when it moved from the FTP site:

1. A dialog appears with the title Moving... and it says it's calculating how much time it'll take. This took a while itself.
2. Then the dialog says it's moving the files and the progress meter shows the progress
3. After the progress meter fills up, it goes to empty again and then says it's deleting the files it moved

I did this with most of the folders until I reached one that had about 18GB of data in it. I performed the same procedure as above to start the move and after it was copying files for about a minute or two I realized that I was moving them to the wrong location (my desktop instead of a directory on my C drive). At work our desktops are retargeted to a file server and not our local hard drives. I knew that by copying to the desktop I would hit my quota limit on the server whereas if I had copied to the C drive like I did with the other folders I had no limit aside from the free hard drive space. Not copying there in the first place was just a mistake. I clicked cancel to stop the transfer. When I did that, the dialog that had the progress bar changed to the empty progress bar and said it was now deleting. By the time I realized what was going on, it had deleted that folder and all of its content from the FTP site.

Now this could have been some weird bug or interaction between the fact that I was using a machine with SP1 instead of SP2, that my desktop and profile were retargeted to another machine, or that I was moving so much data. It wasn't a lot of files as these were data files for desktop publish programs for some brochures and catalogs, along with large print-ready images. I don't know what items worked together to cause the problem. I do know that it did happen and that I had to deal with the IT group at our other site (that had the Solaris server), open the helpdesk ticket to get them to restore the files from backup, wait for them to get around to it, etc. It was a huge pain and delayed what I was working on, causing grief for myself and my internal customer. One thing's for sure; I'll never use the built-in Windows FTP client again.

Re:That's silly.

[Skuld-Chan]

Isn't this a Mac? I thought they were supposed to be easy to use.

Re:That's silly.

[kisielk]

There have been resource fork patches for some time, but somewhat unreliable. Version 3.0 is going to support resource forks (and other types of extended attributes) out of the box. The setup we have been using extracts resource forks to a separate file on the mac and then backs them up alongside the original files. The restore process just performs the reverse. It will be nice once we can switch to rsync 3.0 and get rid of that step.

Re:That's silly.

[delire]

resource fork... from Apple's rsync manpage: -E --extended-attributes copy extended attributes, resource forks

With cryptic commands just to copy files Apple will ALWAYS have ~.001% market share. When will Apple get it into their head that programmers shouldn't design user interfaces?

Why doesn't Apple just copy the way Ubuntu does it and get with the 2000's?

Pride is what it is. That damned pride!

Tiger has this problem as well!!!

[itsybitsy]

I lost a huge amount of data being MOVED (not copied) from one volume into a virtual volume DMG file. Lost and gone forever, lots of important files. What happened? I simply bumped the laptop Mac Book Pro during the move... zap... gone forever. The DMG file was blank! Yes, complely zero bytes except for a bit of header non-file data. It sucks bad.

Re:Tiger has this problem as well!!!

[Chouonsoku]

Just wanted to also confirm that the bug was in Tiger. I was backing up music files to do a clean format for Leopard and lost everything when the hard drive got disconnected by mistake.

Re:Tiger has this problem as well!!!

[Liquidrage]

Can't agree with you. It is the exact type of that is usually caught by automated testing. The issue isn't that a hard drive was bumped. The issue is that the write operation failed. In this case due to a drive no longer being accessible. The failure is easily automated, and the result of that failure is easy to catch.

And I wouldn't exactly call this regression testing, as such functions as file movement aren't usually impacted by later changes. It should be pretty basic on the design chart. Sounds to me more like "working as intended...use move at your own risk". Which I think it stupid, but I don't see how this really was *missed*, especially since some are claiming it's been this way since at least Tiger.

Re:Tiger has this problem as well!!!

[Otter]

I've had that also (on Jaguar, IIRC) when an external drive lost power. Lost both the data and the old backup, simultaneously.

Re:Tiger has this problem as well!!!

[dal20402]

So your irreplaceable videos were made the previous year and you had no backup.

I suppose the PC makers should like you. You'll shoot up a PC and have to buy a new one every couple years. And each maker will get a piece of the action, since you'll blame the previous maker for routine data loss.

Or you could just do what everyone in the computer business has been telling you to do for at least 30 years and... make a backup.

Re:Tiger has this problem as well!!!

[IceCreamGuy]

Dude, Mac users are just getting this kind of functionality? Come on, we've had that feature in Windows for years...

A great disturbance in the Apple

[syylk]

...As if millions of fanboys suddenly cried out in terror and were suddenly silenced.

Re:A great disturbance in the Apple

[kerohazel]

Yes, but if their data is stricken down, it will become more powerful than we can possibly imagine.

Re:A great disturbance in the Apple

[ILuvRamen]

In memory of them, I'm officially nicknaming Leopard: Vistapard.

Re:A great disturbance in the Apple

[Peter La Casse]

In memory of them, I'm officially nicknaming Leopard: Vistapard.

"Vistard"

Wait...

[ackthpt]

It's kind of strange that this didn't come up while people were beta testing OS X 10.5. Samba is used in many places. I hope they get it fixed soon.

Exactly which decade did you fall into your recently awkened from coma in? Testing? Testing? Nobody tests anything anymore, they just go play with all the new toys and stare at the eye-candy. Actual mundane, humdrum testing? That's an SEP if ever I didn't see one.

Re:Wierd

[cnettel]

All fanboys were just happy with how blazing fast file copy was compared to Vista. The non-fanboys tried to check the real size of the dir by copying it to a Vista machine afterwards, but the progress bar got stuck on 413 hours left and counting, so they couldn't file the bug in time.

You just don't get it...

[juanfgs]

This is the new Leopard "iLostMyFrigginFiles" feature, next version they will add a badass black hole effect when it does that!

Re:You just don't get it...

[DigiShaman]

Oh great! Soon, Windows users will experience "white holes" where Mac files magically appear in the My Documents folder.

Re:Terrible bug

[Liquidrage]

Yes, that's easily avoidable. If you know it exists. And hopefully the first time you get bitten by this it isn't something critical. Would be garbage to have to work around that one.

TFA looked decent in it's details. Even step by step recreation. But it's a pretty serious bug, that as you mention, *needs* to be fixed quick and I didn't see any other sources confirming it.

Par for the course?

[GoRK]

No offense meant here, but normal move/copy operations are traditionally highly destructive events on MacOS anyway. For instance there is absolutely no simple way to merge two folders contents together on the mac. If you drag a folder called "Documents" into your home directory and click on "OK", the Mac OS will happily delete your entire documents folder. I was reminded of this enormous frustration while recovering from some multi-volume backups recently, having to resort to an obscure OS X commandline tool 'pax' and Leopard's newfound support of hardlinks to make some simple file copies play nice and not unnecessarily consume 3 times the disk space they should have.

For all of the flack the Windows file copy interface gets, it is both safer and more flexible than trying to use the Finder: an interface that makes file management so stupefying it becomes impossible.

Re:Par for the course?

[Blakey Rat]

That's left over from the original spatial Finder design in Mac Classic. Apple hasn't really decided whether they want to get rid of the spatial interface, so instead they've made this horrible frankenstein half-spatial, half-browser interface which pretty much everybody hates.

Doing a "replace" for that operation makes sense in a spatial system because all spatial icons are treated the same way. You'd wouldn't expect dragging a Word file named "happy.doc" into a folder already containing a "happy.doc" to perform a merge operation; so why would you expect that with a folder in the same situation?

That said, if you've never used Mac Classic, you'd think OS X has nothing but a browser interface, in which case all metaphors and ideals are out the damned window, and the OS might as well do a merge operation. Since you most likely came from Windows, or a Linux environment ripped-off from Linux, you'd expect dragging identically-named folders together to do a merge operation because that's what you're used to.

Apple needs to make up its mind what Finder is. It gets worse and worse every version.

Re:Par for the course?

[JunoonX]

When two folders, both named "Documents", where one is dragged and dropped into the home directory containing another "Documents" folder, Windows prompts if you want to replace content from the dropped folder on to the one being dropped on. At this point, if any files with same name are encountered, they will be replaced with the one from the first directory; however, all other files in folder will stay intact.

Re:Par for the course?

[nine-times]

If you drag a folder called "Documents" into your home directory and click on "OK",

To be fair, I don't think it asks you whether it's ok to move that directory. It will warn you that it's going to replace that folder, and the buttons will either say, "Replace" or "Stop". It's not that ambiguous.

The only thing that makes it problematic is if you're accustomed to working in a file manager that will automatically merge directories, then you might think it's going to merge when it's actually going to replace. I would say that neither behavior is "wrong", but you certainly can get unhappy results if you're expecting one behavior and get another.

Honestly, it took me a little while to get used to it, but now that I expect it, it's fine. Usually, if I'm doing anything complicated with copying/moving lots of stuff recursively, I'm going to want to use a command line anyhow. In the command-line, "cp" and "mv" work in normal unix fashion.

Re:Par for the course?

[node 3]

For instance there is absolutely no simple way to merge two folders contents together on the mac. Open a folder, "Select All", drag into destination folder.

If you drag a folder called "Documents" into your home directory and click on "OK", the Mac OS will happily delete your entire documents folder. It does state this is going to happen in the window with the OK button.

I was reminded of this enormous frustration while recovering from some multi-volume backups recently, having to resort to an obscure OS X commandline tool 'pax' and Leopard's newfound support of hardlinks to make some simple file copies play nice and not unnecessarily consume 3 times the disk space they should have. If you were going with the command line, you could have just used "cp" or "rsync". Your mention of hardlinks is perplexing, though. OS X has supported hard links forever. It recently added support for hard linking folders (extending hard links beyond the standard). Since you're comparing with Windows, does Windows somehow know if you are copying a file that's identical to one that already exists, and makes a hard link? I'm fairly sure it doesn't.

For all of the flack the Windows file copy interface gets, it is both safer and more flexible than trying to use the Finder How so? This bug aside, I don't see how it's safer *or* more flexible. The difference--the *only* difference--here is whether folders replace or merge. Windows isn't more flexible, as it does one way, not both. As for safer, they both tell you when something destructive is going to happen.

the Finder: an interface that makes file management so stupefying it becomes impossible. Impossible? Really? All because it replaces folders (and tells you, with a chance to abort) instead of merging them?

Re:Par for the course?

[mctk]

or a Linux environment ripped-off from Linux

Forking Linux developers!

Re:Par for the course?

[TheNetAvenger]

I haven't noticed different behavior in any version of Windows. How do you merge similarly named folders during a copy/move in Windows? In my experience you get the same "Do you want to replace this?" type prompt you get in OS X.


Serious answer...

XP offers a basic, do you want to replace folder and a basic do you want to replace files. (Very generic, but more than OS X does)

Vista on the other hand, asks if you want to replace/merge folder and then if conflicting files are found in the folders it asks you if you want to overwrite the files, don't overwrite them, or create a second copy of the file in the same folder. File by File and Folder by Folder if you want, or you can hit a check box to mimic your response for all file and folders if you are moving a lot of files/folders.

It also will show you the two versions of the file or folder so you can make a decision based on which files/folders are newer, and you even get a thumbnail of the file for documents and pictures to be sure you are keeping the one you want.

(Very simple interface, but more has the functionality of the power geek that was always left to using tools like XCopy in the past.)

This is one of the 'little' Vista features that doesn't get any press, but is a lifesaver for people that move around a lot of data, as you can merge and update folders and files much easier.

Stuff like this is the reason I said if MS did a 'new features' list like the pety list Apple did with their 300 new features, Vista would have several thousand new features to list.

(Again MS's marketing sucks, since most people don't even know stuff like this exists in Vista, and it is both powerful, easy, and just works.)

Re:Par for the course?

[TheNetAvenger]

Basically, it nukes the conflicting /alpha/baz and then performs the copy. Brain dead behavior if you ask me.

Vista works very differently, not only Folders are confirmed but every file in the folders are confirmed with options to replace/don't copy/create second copy...

So ya this sucked in older versions of Windows, but Vista does a great job of handling this, better than any other GUI file manager I have ever seen to date.

Maybe you should try Vista instead of OSX... But again if MS was doing their job and was touting features like this in Vista, people would find more reasons why Vista has a lot of things to offer. Instead MS's marketing is retarded and nice features like this are NEVER mentioned and none of even the tech press notices them or points them out to users.

And trust me when I say there are literally 1000s of features like this that make the difference between Vista and XP night and day for daily usage.

Re:Terrible bug

[thegrassyknowl]

Another thought springs to mind... is this even that critical? Doesn't leopard have the time machine in it? Can't you just go back and get your files out of the time machine if they were that important?

I haven't "upgraded" yet so I don't actually know much about this Time Machine thing and how it works.

Take advantage of Time Machine

[tji]

Not to be glib, but.. This would be a great demonstration of the value of "Time Machine" backups. Time Machine is not perfect, but it is a good start on a backup system well integrated into the OS. The example problem, data loss, would be really easily recovered via Time Machine.

Beyond the basics that every decent backup app does, the things I like about Time Machine are:

- Integration into Applications. For example: "Show me what my iTunes library or iPhoto library looked like last Thursday"

- Integration into OS install. In the case of disk failure, recovery to previous state is simple - rather than multi-step with a separate backup app.

Some things that need improving:

- Better handling of file exceptions. I keep work data in encrypted disk volumes (DMGs). If I change one byte, the whole huge file needs to be backed up as each change is detected (generating MANY copies of that big DMG). The only other choice is to say "ignore this file/directory". Same thing applies to any large file, such as a VMware VM file. A better option would be to say "Back this file up, but only keep 'n' versions".

- Time Machine has gotten twice, pegging the CPU/fans on my MacBook Pro.

Wow

[Zebra_X]

Unbelieveable. Forgot to check the result of the copy operation eh? So basically this is a catastropic defect for people who deal with very large media files to and from remote stores or people who deal with virtual machine images.

Back in the day when I used to use my mac I dropped a directory (A) into another directory (B) but there was an existing directory (C) with the same name as (A). The finder asked me something, I clicked OK. I was dismayed to find that the dialog had asked me "Would you like to replace directory C, with A?" - Why on earth would that ever be the default option for a directory move? From the users perspective you aren't really moving the directory, the intention is to move the files, thus the sane response would be to merge A with C not replace it.

Whatever.

Ah, the "outsourcing" coding model..

[eniac42]

Advert on Amazon Mechanical Turk:
Write OS-X compatible application to Move a file between two filesystem devices..
Time Allotted:: 6 hours. Reward: $10.00..

Re:Ah, the "outsourcing" coding model..

[Larry Lightbulb]

rm $from Now where is my $5?

Re:Ah, the "outsourcing" coding model..

[beav007]

Um... no. REALLY no. Please don't do that.

The syntax for "||" is:
If command 1 fails, do command 2 - otherwise exit (where you used "command1 || command2").

In this case, your command will either copy all the files from $from to $to or delete all files at $from.

What you probably meant is cp $from $to && rm -r $from, which only performs the second command if the first succeeded. This solution is far from perfect for reasons mentioned by other posters, but it's still significantly safer and more useful than yours...

Re:Ah, the "outsourcing" coding model..

[megaditto]

And that's probably why you don't work for Apple!

Re:Ah, the "outsourcing" coding model..

[space_in_your_face]

But that's exactly what the Leopard Finder is doing!
Not exactly. The Leopard finder is doing something like:

cp "$from" "$to" ; rm -rf "$from"

like in "remove $from regardless if the copy was successful or not".

Re:Terrible bug

[El Lobo]

but I'm sure Apple will have this fixed pretty quick.

Well, the problem is, this bug exists even in TIGER and has been repported many times! And no, not fixed yet. Abble is a coorporation like any other, and not the superpower that some users seem to think they are.

Re:Terrible bug

[Knara]

You're asking if a bug wherein entire folder hierarchies can go *poof* in the event a network share drops should be considered critical? Are you serious?

Apple Customer Quality Feedback (CQF) Program

[SinceEBCDIC]

If it's true that Apple's products seem less-tested these days, it's because they've tossed lots of seasoned customer quality-assurance testers to the wayside.

Many years ago, think System 7, Apple had this great Customer Quality Feedback (CQF) program. We tortured our systems between the alpha-testers and the great unwashed masses. There were Apple staff who (gasp) listened to our bug reports and got back to us reasonably quickly. It was grand.

Then someone got fired, or promoted, or whatever, and the CQF program got lost in the shuffle. Every few years I get an email from someone at Apple telling me that they're reconstituting it, but nothing ever comes of it, and - you know - it's hard to understand how they can ignore a free, highly-motivated bunch of fanboys.

"haha"

[MutantEnemy]

Why is every destructive computer bug that happens tagged with "haha"?

Data Loss Bug In OS X 10.5 Leopard
bug, macosx, apple, haha

Symantec Updates Cause Chaos in China
haha, security, bug, windows, feature

Banner Ad on Myspace Serves Adware to 1 Million
haha, myspace, pwnd, security, adware

Ubuntu May Be Killing Your Laptop's Hard Drive
linux, haha, storage, bug, spam

Islamists exploit buffer overflow, hack U.S. nuclear command; world doomed
eschaton, religion, waronterror, haha

OK, I made one of those up. But it doesn't even matter what OS or company is responsible for the problem - whoever makes the tags seems to take great delight in all computer snafus. How does the tagging system work anyway? It's always been mysterious to me.

Re:I don't understand

[myowntrueself]

I agree. I always copy then delete, especially when dealing with network shares.

Despite its many shortcomings, Windows ME (*NOT* the more recent ME2) had this truly wonderful feature where if you delete something from a network share it would *copy* the data across the network into your trash folder.

Really handy when you delete 10G of data on a network share and your local hard drive has 5G available and you are on a 10mbps network.

Not default behavior

[Have Blue]

Someone should point out that by default, the Finder will copy files dragged to a different volume. You have to hold down a modifier (Command) to force a move operation. So while this is a severe bug that should be fixed immediately, it's much less likely to happen by accident than the article implies.

Re:I don't understand

[CaseyB]

I call bullshit on Windows and Linux, and I'm pretty sure you're wrong even for DOS.

Maybe move should be implemented as copy, completly then delete but its often not.

Why on earth wouldn't you? Doing it any other way is not only obviously dangerous, it's far harder to implement! What would you do, map the file on disk, unlink the file and then copy and wipe each raw disk block? I could see doing something crazy like this in specialized applications, where freeing storage at the source is priority one, but in the general case it's insane.

The Apple issue is clearly just a stupid exception handling bug.

Re:I don't understand

[t35t0r]

NFS write failure on Linux 2.4, check your data is gone.

Uhh no. In linux mv's always make sure the data moved then delete the original file (as far back as 2.4). You can test this by dd'ing a large file (use if=/dev/urandom), then run a md5sum on it, then do a mv within the same drive, over nfs, over samba/cifs, to floppy (not sure what happens here because linux caches writes to floppies until umount), to usb drive, whatever. While the mv is in operation just pull the plug on the system (target or source system). Your old file will still be there!

Re:No it isn't

[tf23]

You have to pay to be able to participate in the program

Wrong-o!! You absolutely do not have to pay anything to involved in the Appleseed program.

Re:Terrible bug

[TheNetAvenger]

is this even that critical? Doesn't leopard have the time machine in it? Can't you just go back and get your files out of the time machine if they were that important?


Only if a backup of the files was run, which is a requirement of Time Machine.

If this was Vista, then there would be a good chance to use 'previous versions' to recover the folder data, as it does not 'solely' rely on external backup for timeline file recovery. Vistas use volume level file version snapshots(a feature of NTFS that HPS+ doesn't support), so there are backups even on the drive if it hasn't ever been backed up.

(Remember Time Machine usually only runs once an hour, and all versioning or changes made in that hour are never kept or tracked.)

-PS Not trying to troll, but this is a perfect example of the difference between Vista's previous versions and Apple's Time Machine I have tried to point out in the past.

Vista does both volume level backups and external backups, unlike just external backups like Time Machine does.

See why IT people prefer Vista's method, even if a backup hasn't run that hour, even users themseleves can access files and folders rather easily that got deleted or changed. And since this has been around Since Windows 2003 Server, in corporate environments, even XP users accessing 2003 servers have had this feature for over 4 years now.

Vista's claim to fame is that it enables these features on the local hard drive, and also integrates with the Vista backup system, so file versions appear from the backups in addition to the 'snapshot versions' on the main hard drive.

File The Bug!!

[EccentricAnomaly]

Go Here: http://developer.apple.com/bugreporter/ And File a bug. And They'll fix it.

Re:Terrible bug

[Hierarch]

Did you miss the part where Leopard reported an error on the copy? (No, really... I'm not being snarky, did you actually miss that screenshot?) Leopard correctly recognized that the transfer had failed, and still deleted the source.

To me, that's a pretty clear bug. What stuns me is that Apple isn't using the underlying "mv" command - since it should certainly deal with this situation. They rolled their own defective version.

Never re-invent the wheel. You'll have a square wheel with 13 lug nuts of all different sizes. Just go to Goodyear and take the tried-and-true.

That's silly? You're silly.

[Nirvelli]

There's absolutely no reason that a modern "move" command shouldn't already be implementing the exact process you've just described, no matter where it is moving to.

Re:defectivebydesign

[klubar]

You have to word the dialog box more precisely. "Do you wish to not allow data corruption or success on loss of destination meda?" Cancel? Retry? Abort? Ok?

This may not be a bug in Leopard

[ChrisA90278]

There is a good chance the disk drive or the networked computer is giving the finder an "OK I got it" signal BEFORE the data is actually written to the physical disk drive. So it gives this signal, then the finder deletes the old data and then the disk drive "goes away" before the data is written out of the cache and to the drive. Don't blame Finder, blame those huge disk caches where many megabytes sit and wait to be written to the drive.

If this is true then the bug is im the copy of Samba running on the file server. We do not yet have enough information to know.

Re:This may not be a bug in Leopard

[Cantus]

The default Finder behavior when you drag a file from one volume to another is to copy the file. To do a move you press the Command key and drag the file to its new destination.

The second you drop the file, it begins copying the file to the destination volume, but the original file disappears... poof, gone. If you stop the move operation, you're left with an incomplete file in the destination volume and with nothing in the original volume. This is to me a major bug.

This is why, for large files, I always copy.

I had this happening to me with WinXP aswell

[Lisandro]

That's it, moving a file, network failure and poof, moved files deleted from the source. Most probably a bizarre coincidence, but after it happened a few times i now copy and then delete when working on a SMB network.

(Or maybe Samba has something to do with it?)

Re:Terrible bug

[RobertM1968]

Error code 51 doesnt neccesarily mean the transfer failed. It can also mean lost connection to the share - which could have happened (in OSX's "mind") after it thought/was told the transfer was complete.

For instance, what if in stopping the share on the Windows session (incorrectly listed in the /. article as a FreeBSD Share), the SMB crap in Windows is "cleaning up" while shutting down and generating a "complete" code that it sends back to the Apple machine, which then in trying to communicate with the share, realizes it no longer can, and generates the correct error code as noted (51).

OSX did NOT generate a disk write error or any other error that would have been more applicable (like 43 or a bunch of other possibilities). It generated an error stating it lost communication with the other machine.

Again, still not enough info without knowing more about what the WinShare does when closed.

Re:Problems

[Space cowboy]

No need to mod you down. You're just wrong.

simon% dd if=/dev/zero of=test bs=1m count=17
17+0 records in
17+0 records out
17825792 bytes transferred in 0.297437 secs (59931329 bytes/sec)
 
simon% ls -l test
-rw-r--r-- 1 simon 501 17825792 Nov 5 18:32 test
 
simon% /usr/bin/time cp test ~/Desktop
        0.44 real 0.00 user 0.03 sys
 
simon% ls -l ~/Desktop/test
-rw-r--r-- 1 simon 501 17825792 Nov 5 18:33 /Volumes/Users/Users/simon/Desktop/test
 
  simon% uname -a
Darwin mac 9.0.0 Darwin Kernel Version 9.0.0: Tue Oct 9 21:35:55 PDT 2007; root:xnu-1228~1/RELEASE_I386 i386

... I suspect you have a hardware problem if it's really taking that long to copy files. The above 0.44 secs (wall-clock time) is on a standard internal SATA disk. No RAID or anything special.

Simon.

Temporal problem

[Roger W Moore]

In any case, you should be using Shadow Copy...er...Time Machine which would have protected you from going and losing track of your own files.

Great! So now not only don't we know where our data is but when it is. Perhaps in a week or two's time the data will materialize in the folder it was supposed to be moved to with an accompanying "whorping" sound coming from the speakers?

That's missing the point...

[megaduck]

Speaking as one of those IT people, NTFS is probably one of the coolest pieces of software ever to come out of Redmond. ACLs, alternate data streams, directory junctions, single-instance stores, shadow copy, the list of useful features is huge. Even more surprisingly, it works pretty much as advertised. Frickin' cool.

There's another angle, though. On paper, Vista's NTFS-based backup technology walks all over Time Machine. However, the USABILITY of Vista's technology is crap. This morning, I enabled Time Machine by plugging in a USB drive and clicking "Use as Backup Disk" when prompted. To do restores, I launch the cleverly named "Time Machine" application. I've already used it twice today just because it's fun to watch the spacey animations.

Compare that to Vista's clunky "Backup and Restore Center", which you have to use if you want to backup your files on an alternate volume. I guarantee you that using "Backup and Restore Center" is beyond most average users. Sure, it might be "better", but what good is it if it never gets used?

Re:Problems

[Daengbo]

How can you have a 5-digit UID and still have never seen the 17MB troll? Did you buy your account on ebay? P200. How old do you think the troll is?

Re:Or [possibly], go fix it.

[tftp]

Apple charges good money for Leopard, so let them fix it.

Re:Or [possibly], go fix it.

[andreyw]

Finder, and any GUI parts of OS X in general, are not part of Darwin.

Re:No it isn't

[tftp]

Someone already mentioned that the bug is not in copy or delete functions of the kernel - it is in the Finder GUI that is not open source.

Re:No it isn't

[JEB_eWEEK]

Not quite.

How can I sign up for this program?

We are not seeking new participants. You must have an invitation in order to log in.

https://appleseed.apple.com/cgi-bin/WebObjects/SeedPortal.woa/wa/faq

OR, there's:

Become an ADC Select Member US$499.00

Upon purchase and activation, you'll receive:

* ADC Software Seeding Program

http://developer.apple.com/products/select.html

Re:Terrible bug

[Pliep]

Actually, Time Machine runs every two minutes. You can see a countdown timer in the system preferences. Also, Time Machine is dodgy by itself. Apple confirms that after backing up about 10 GB of data, Time Machine may stop working. See here: http://docs.info.apple.com/article.html?artnum=306932

This may not get read, but still

[analog_line]

This is a bug with the FINDER ONLY, just so everyone is clear. The Unix "mv" command in Leopard is NOT affected by this.

http://www.macintouch.com Obviously this is the front page story today.

This only affects "move" operations between logical volumes. You have to hold down Command while you're doing this, inside the Finder, to get this to happen. Yes, it's a bad bug. It's not something, however, that you're going to run into if you're thinking sensibly about how you treat important data, or if you didn't know that the Command-drag functionality was built into the Finder (which I didn't, and I can't think of a time when I would use it now that I do, even if it was working correctly).