Massive data loss bug
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. Here's the link - http://tomkarpik.com/articles/massive-data-loss-bug-in-leopard/ I hope Apple would get this fixed soon.
That is actually the safest and easiest option.
Seems easier to command-click drag and drop to me. Losing data is not acceptable under any circumstance.
Everything has been precisely calculated
I don't think that data loss is something that was purposefully precisely calculated.
A short term fix would be for Apple to disable the command-drag n' drop altogether, long term would be to fix the root issue in the first place.
Am shocked to hear that this data loss bug has been known about since 10.3.
Similar Messages
-
Serious data loss bug remains!!
I had filed a bug report on this for v1, but there appears to be no change in Pages version 2.
Pages has no problem allowing multiple people to open a file on a server at the same time. Each time any person hits save, their version of the document replaces the current version (which may contain changes made by others who have the document open). No warning is given to any party.
This makes it very hard for multiple people to collaborate on a set of shared documents, as they won't know who might already be editing the document.
This seems glaring and dangerous behavior.
Does anyone know a good LIGHTWEIGHT check in, check out system that can be used to work around this?
-PrestonAs I recall, no where in the Pages documentation or on the website does it indicate that there is a collaboration system in place to allow what you are doing.
Pages let's one person work on a document at any one time One person opens, and one person can save. If you have opened it you can save it. If someone else has opened it in the meantime, they can save it too. If one overwrites the work of another, that is indemic to a single-user-oriented document and program.
What you are doing is like five people trying to steer a vespa, and yet only one person can actually sit on the bike.
It's not a bug, but by all means do send feedback to request a collaboration feature or proper versioning system for netowrk saving be implemented in the future. -
Warning: SERIOUS Data Loss Bug in 10.5.2 Time Machine
I've been backing up an aluminum 10.5.1 iMac and two other computers to a network drive on a 10.5.1 MDD G4 for several months. After updating both machines to 10.5.2 this morning, Time Machine deleted ALL of my backups for the iMac. system.log tells the sad story:
Feb 12 10:06:25 tim /System/Library/CoreServices/backupd[429]: Backup requested due to disk attach
Feb 12 10:06:25 tim /System/Library/CoreServices/backupd[429]: Starting standard backup
Feb 12 10:06:25 tim /System/Library/CoreServices/backupd[429]: Mounting disk image /Volumes/Time Machine HD/Tim_001b639d09d8.sparsebundle
Feb 12 10:06:26 tim hdiejectd[436]: running
Feb 12 10:06:27 tim /System/Library/CoreServices/backupd[429]: Disk image mounted at: /Volumes/Backup of Tim
Feb 12 10:06:27 tim /System/Library/CoreServices/backupd[429]: Backing up to: /Volumes/Backup of Tim/Backups.backupdb
Feb 12 10:06:27 tim fseventsd[41]: bumping event counter to: 0x3bb17fc (current 0x314c) from log file '0000000003bacd10'
Feb 12 10:06:31 tim /System/Library/CoreServices/backupd[429]: Event store UUIDs don't match for volume: Macintosh HD
Feb 12 10:06:31 tim /System/Library/CoreServices/backupd[429]: Forcing deep traversal on source: "Macintosh HD" (mount: '/' fsUUID: 3CC031CF-68B8-387F-ADB8-DC40D76C8806 eventDBUUID: AE82F29F-20AC-4C31-8E52-4CCA8D4E37E5), {1, 1, 0, 1}
Feb 12 10:06:31 tim /System/Library/CoreServices/backupd[429]: Node requires deep traversal:/ reason:kFSEDBEventFlagMustScanSubDirs|kFSEDBEventFlagReasonForceDeepBackup|
Feb 12 10:06:45 tim /System/Library/CoreServices/backupd[429]: Backup requested due to disk attach
Feb 12 10:20:59 tim /System/Library/CoreServices/backupd[429]: Starting pre-backup thinning: 3.43 GB requested (including padding), 2.61 GB available
Feb 12 10:21:16 tim /System/Library/CoreServices/backupd[429]: Deleted backup /Volumes/Backup of Tim/Backups.backupdb/Tim/2008-02-11-094615: 2.61 GB now available
Feb 12 10:21:27 tim /System/Library/CoreServices/backupd[429]: Deleted backup /Volumes/Backup of Tim/Backups.backupdb/Tim/2008-02-11-084618: 2.61 GB now available
Feb 12 10:21:27 tim /System/Library/CoreServices/backupd[429]: Removed all 2 expired backups, more space is needed - deleting oldest backups to make room
... Dozens more deletions removed ...
Feb 12 10:55:11 tim /System/Library/CoreServices/backupd[429]: Deleted backup /Volumes/Backup of Tim/Backups.backupdb/Tim/2008-02-12-054723: 2.61 GB now available
Feb 12 10:55:26 tim /System/Library/CoreServices/backupd[429]: Deleted backup /Volumes/Backup of Tim/Backups.backupdb/Tim/2008-02-12-064614: 2.61 GB now available
Feb 12 10:55:43 tim /System/Library/CoreServices/backupd[429]: Deleted backup /Volumes/Backup of Tim/Backups.backupdb/Tim/2008-02-12-074612: 2.61 GB now available
Feb 12 10:55:43 tim /System/Library/CoreServices/backupd[429]: Error: backup disk is full - all 59 possible backups were removed, but space is still needed.
Feb 12 10:55:43 tim /System/Library/CoreServices/backupd[429]: Backup Failed: unable to free 3.43 GB needed space
Feb 12 10:55:43 tim /System/Library/CoreServices/backupd[429]: Backup failed with error: Not enough available disk space on the target volume.
Feb 12 10:55:50 tim /System/Library/CoreServices/backupd[429]: Ejected Time Machine disk image.
Feb 12 10:55:50 tim kernel[0]: AFP_VFS afpfs_unmount: /Volumes/Time Machine HD, flags 0, pid 502
Feb 12 10:55:50 tim /System/Library/CoreServices/backupd[429]: Ejected Time Machine network volume.
The sparsebundle was over 100GB, even after Time Machine erased everything. My best guess is that Time Machine isn't compacting the sparsebundle after it deletes, and is checking the network drive's free space rather than the sparsebundle's free space, leading TM to delete everything in the bundle in an attempt to free more space.
Message was edited by: tgildeaFileVault isn't involved here. Time Machine uses sparsebundles when backing up to a network share on another Mac.
In regards to the UUID: I didn't do anything that should cause this to change. Looking over system.log some more, this message didn't appear until after I installed 10.5.2. Time Machine was working properly just minutes before doing the update.
Message was edited by: tgildea -
Thank-you in advance for reading all of this.
A slip of the mouse in trying to return to the left-most tab caused the tab groups feature to take over the window. I completely failed to comprehend what I was looking at and closed the thing by clicking the x because I damn sure didn't want what I was looking at. Undo close group? Of course not, I didn't want a group in the first place. Hey! My tabs are completely gone now! So in just a few steps all of which seemed innocent, I have lost all but ten of the tabs that were open, and those were only the most recent ten (and actually older tabs would have been more valueable than newer so insult on top of injury).
I don't think increasing the number of undoable tab closures is the answer. I think getting rid of the tab groups feature all together is the best way to go. I seem to have lost all the tabs I had open, irretrievably, and some of them had been open for WEEKS so they can't be reopened from history.
The tab groups feature is a data loss bug. Please remove it from the UX completely. If you have to figure out a way to grandfather people who actually have a tab group, fine but don't further vex me with it, please.
The reason this catastrophe has occurred is that your feature, was so discoverable that I activated it by mistake, and the UI after that had to be learned on the spot, yet was too complicated to learn. You really should withdraw it until you can figure out how to keep the UI for it from causing data loss such as I have experienced this morning.
Also, there may be other questions on this feature which are from a certain point of view the same as this one. It's hard for me to tell for sure. None of them had a true resolution so that's why I posted this one, also I'm seeking a specific fix, removal of the offending feature.
Why remove the feature? Because if I allow in my discussion the possibility of retaining it, it will stick around in its present and dangerous form until your team is done bikeshedding the new form it should take. Instead I want you to delete the feature and then require its champion to convince you that a new incarnation won't cause data loss before you allow it back in.
Fellow users, if you know of a way to recover all 50 lost tabs from an "undo close tab group incident", I would be grateful for that advice, but I would still believe the feature should be deleted. Some of the tabs have been retained for months so digging them out of the history is not practical.I am not 100% certain but I am fairly sure development of this tab feature has ceased and it is likely to be deprecated. I am even less sure what the replacement will be.
Thus forum is not the place for discussing development issues or feature requests. (Developers do not use this site)
*You could comment using feedback <br /> https://input.mozilla.org/en-US/feedback
*Or try to find a developers forum or mailing list <br /> http://www.mozilla.org/about/forums/#general-development
I have not tried this and am not even using Windows at present but the tabs groups are stored as entries in the sessionstore.js files. Firefox has such a file plus sometimes numbered copies and a backup copy. Firefox will have re-writen these files but I suppose there is a posibility the Windows OS may have kept copies.
* see http://windows.microsoft.com/en-US/windows7/Previous-versions-of-files-frequently-asked-questions
* http://kb.mozillazine.org/Sessionstore.js Note the files are kept within the Firefox Profile
** Firefox Button -> Help -> Troubleshooting [open profile] (or some thing similar) is one method of locating the profile.
Find and backup any current sessionstore.js file. Consider also bookmark or otherwise recording anything you have open. Now find any previous versions and try putting them in Firefox's profile before restarting it. -
Chronic catastrophic data loss with Lion & servers
Auto-save and versions, when combined with multiple users and a server environment virtually guarantees catastrophic data loss.
Here's how you replicate it.
(1) Place a photo on a server.
For us, this is where we recommend people keep important files. Their desktop may be backed up with time machine, but the server is RAID 6 with hot spares, and on top of that is backed up every night. Much, much safer.
(2) Open that photo in preview. Crop it so that you can print it. Close preview. Notice it didn't ask you if you wanted to save?
(3) Open that file again. It's still cropped. The original? Gone. Choose "Revert to Saved" from the file menu. Notice the large number of empty windows on the right? Yup, you can't get back to the prior version, even though you never saved.
Now, as a network administrator, imagine the consequences of this for your network. Users accessing shared files and making changes they need, but not saving them because they don't need to be saved. Yup, their version is now THE definitive version for the entire network.
The ability to do non-destructive edits is something that computer users have been doing since, well, the dawn of computing more or less. Lion removes this. Now, add on top of that user mistakes in a network environment. User accidentally deletes all the text from a 24 page document. They don't know what to do so they close the app. Unless the backup has run since that 24 page document was created, it's gone. Forever. And ever. Amen.
Go to backups you say? What if the document isn't accessed for six months from when the user accidentally wiped it out. Hope your backup server is petabyte big, because otherwise that data is gone.
This is a huge, huge problem.
JoelI am not an idiot. The very reason we are reporting this is that we are testing this and discovered it. It is not about "random people" changing things on the server. People with appropriate access can do catastrophic damage without even realizing it. Yes, I have also reported this via the Lion feedback.
Let me address some of the comments here:
> There is nothing to prevent someone on Snow Leopard from doing the same thing if anyone can save a document.
That's the problem. Nobody has to actually save to wipe data out with Lion. Versions _does not work_ on server volumes. I have tested this with Lion client and 10.6 server and 10.7 server. Same result. Massive data loss.
> Plus, any document over two weeks old would pop-up a big "locked" dialog if the user tried to modify it in Lion. You wouldn't get that before.
Great. Documents can only be inadvertently destroyed for two weeks. Not sure how this makes anything better.
>And why do you have Time Machine on the desktop instead of on the Server? It seems like it would more handy there.
The server is backed up, as are the clients. It's irrelevant to the problem being reported.
>You are the network administrator. Disable Preview and get your users a decent image editor.
Great idea! Until Photoshop supports versions. Microsoft has announced support for versions with office. I guess we'll have to disable that as well. Maybe we should try to fix the actual problem rather than just disabling software that used to work perfectly and moving on...
> The Versions feature is absolutely not appropriate for use in a multi-user context.
I agree. So do mac users stop using servers now? Doesn't seem like a good option.
> I find it hard to believe that any competent administrator would 1) upgrade clients to Lion this soon after release
I have no clients except my own on Lion at this point, for testing. This is exactly the kind of problem I need to trace down before I deal with huge issues from my users. Also, what happens when we buy new machines? Lion only...
> and 2) have versions enabled on shared disk.
We can turn off versions on a shared disk? I do not see any way to do this. Please enlighten me!
> 1) There is no evidence that the original poster has actully upgraded any client or is a network administrator.
I am a network administrator. This only happens with Lion and versions. I have tested this problem against both 10.6 server and 10.7 server. It exists on either.
> 2) Version is a feature of Lion, not the network. The scenario described by the original poster could occur on any non-HFS+ disk.
The problem is that versions _does not work_ on an AFP server, but acts as if it does. When you use "revert to save" you find that there are no versions. Not good...
Joel -
Beware of iTunes' Reorganize Library feature - DATA LOSS!
Once again I had to report an INFURIATING DATA LOSS BUG in iTunes to Apple.
I have a large library of music, movies and TV shows-- all painstakingly purchased, ripped and imported onto a 2 TB external Lacie NAS. Everything works fine-- UNTIL I dare to use iTunes' Library -> Reorganize Files feature in an attempt to clean up iTune's inability to keep an external drive properly organized as media is added.
This time, iTunes saw fit to DELETE all of the .m4v files from THREE ENTIRE TV SERIES. With NO WARNING OR ERROR. The files are simply gone off my NAS. These will take weeks to replace!
I reported this bug back in iTunes 10.4 and it STILL EXISTS in iTunes 10.5b48. So beware of using this feature, even in the upcoming version of iTunes.
iTunes should NEVER delete your files without warning no matter what! Not under ANY circumstances!Since this is an iTunes question, it might be better to post in the iTunes for Windows forum.
http://discussions.apple.com/forum.jspa?forumID=800&start=0 -
Data loss Lacie 500GB external drive_ disk utility changed partition!
Details of problem. I have a 500GB Lacie mini hub HDD which is used with an iMac angle poise. Recently it has been playing up- unable to access folders/files etc. and since I know the PSU fails on these items (capacitors fail -5V and 12V go haywire- already fixed another one - I have a 320GB with a mac mini setup) I removed unit from the imac and connected in place of the 320GB lacie on the mac mini, using the known good PSU. The iMac is running latest 10.4 and mac mini is running latest snow leopard 10.6.1
The Lacie spun up and mounted OK and I was able to access all files folders with no issue, so since I was keen to make sure it was 'fixed' I ran disk utility. It ran and said it verified and then I realised I hadn't run repair so re-did this. Big mistake!
It did the same run and said no problems found, but it then said something like updating partition! Horrified I waited and it seemed to finish or had finished straight away. Not sure. I looked at folders and yes main 2 items in place, but nothing showing below this. ****. It got worse. I ejected and tried to reconnect and low and behold it says drive needs formatting now. Disk utility even shows this 500GB as having 2.2TB of space, which I see in other posts is another issue.
I have important data backed up elsewhere, but there are a large number of files I do not have copies of and suspect there is a way to recover these. I have since seen all the 'data loss' guest account issues and wonder if this is yet another bug in snow leopard.
Have tried running Techtool Pro 5.0.6 to see if it can see the data but its not seeing anything. Will try Diskwarrior and others, but may well shelve to see if apple gets an os update released for data loss issues.
Any thoughts on how to recover? Anyone with a similar issue? I am pretty certain this is not a failed hard drive as it sounds 100% normal. Thanks for any help or guidance.Thanks for all the comments recieved. V.K. made me re-evaluate what I had proved or not. I was fairly certain (not 100%) that the drive mech was OK, because the psu had failed, but should have proved drive mech was actually OK. So I removed the mech (as USB abd firewire can't do SMART etc.) and hooked up to an old G4 Powermac. I then found that the SMART verified, but had forgotten the 128GB limitation on older macs. So removed and got a PC to hook it up and run latest IBM_Hitachi drive fitness and other diags. (its a Deskstar HD and I have had almost zero issue with these drives over the years.) It checks out fine, so I was correct in my initial thoughts, only the PSU was faulty.
So back to my initial suspicions that Snow Leopard or disk utility has screwed things up, maybe a bug, maybe not. And I did run verify first and it said it was OK, only when I re-ran the repair did it then do a "partition update" (or whatever it said - should have done a screen capture - searched apple but can't find listing of all mesgs avail. must be in developers site I expect). So I am back at looking at how to recover, if at all possible, but as Alsoft have said this is not something Diskwarrior can do directly. Files must still be there just partiton map is now in a ... needs to be initialised mode. Not that I will do that.
So I guess summary is possible bug/funny in 10.6 and/or disk utility.
sounds like I will be contacting Alsoft now
may read up on partition map damage first, so I get a better picture of whats involved.
Cheers and thanks for all the help/ideas.
BTW - been using macs_pcs for decades now and helping friends & family fix things as and when. This is first time needed to post on forum as always found answers to questions I had previously. Must remember the old adage of not jumping onto new technology until it has bedded down I bet if I had hooked up to the old G4 Powermac I wouldn't be in this position. Thanks again. -
Data loss in the secure file transfer using java
Hi,
My java Appllication uses the sftp(secure file transfer protocol) for file transferring. when the load is high like 150 GB , data loss occuring for some files especially last chunk of data is missing.
Core file transfer code
while (true)
int read = bis.read(buffer); //Reading bytes into the byte array (buffer) from network stream (bis).
if (read==-1) {
break; //Break the loop, if there are no more bytes available in the InputStream.
bos.write(buffer,0,read); //Writing the same number of bytes what we read into buffer byte array into OutputStream
complete.update(buffer,0,read); //Giving the same number of bytes what we read above to the checksum calculation.
bytesSoFar += read; //Incrementing the bytes that we have read from the source.
//if the bytes transferred are equal to the source file size
If (bytesSoFar == fileInfo.getFileSize().intValue())
// Calculate the file checksum.
// Update DbAuditLog with checksum and no. of bytes read
else
// Update DbAuditLog with no. of bytes read
bos.flush();
Please help , why the data loss is happening when the big size files transfer
Note:
1.150 gb not a single file multiple files.
2.SFTP no java specific API, So connection is happening through java .io.sockect connectionso, if you are not getting any exceptions, yet the data is incorrect, it sounds like you are having actual network issues. i'd start looking at things like the actual network cards in the computer as well other equipment involved in the transfer. since you are not getting exceptions, the network stack obviously thinks nothing was lost over the wire, so either the actual hardware has issues, or the network stack, or the issue is on the sending end. the last possibility is that the client is somehow munging the data it is sending. do you control the client code? if so, i'd start checking the client side code for bugs (and/or post it here).
also, i'm assuming that all of the variables included in your original code are thread local (you aren't sharing any of those buffers/streams/etc between multiple threads). -
Data loss in Manage data in HFM 11.1.2. Advice needed
Dear All,
Yesterday while using the Manage data function in HFM we encountered a large amount of data loss. Let me explain.
Our goal was to copy Non Financial accounts Entity Currency data from one budget scenario to another budget scenario. See the below screenshot. While this happened correctly an unintended consequence was that the financial accounts data got erased. So the budgeting team had to end up redoing their budgets again. My question is when we are telling the system to only copy the Non Financial data why would the system even touch the Financial data?? This does not seem right at all.
Later when I was doing testing I changed the above Replace to a Merge and then the system only changed the Non Financial data and kept the Financial data intact. If this is indeed how it works it is very confusing to say the least and does not make sense.
The point being that when we are specifying the account to be replaced or merged why does the system even touch the non-specified accounts? Is this a system bug or its just the way it works? Will appreciate an insight please.Hi there,
Then I will agree with Eric... and I have also just noticed that you have used the Replace option in your screen which definitely has cleared all the accounts before loading the new data.
For me the best approach is to use the clear functionality to delete any previous data from the Non-Fin accounts and then use the Copy functionality with Merge to copy the data.
Regards,
Thanos -
I keep running over my Web host's quotas with my email, so I thought I'd give Gmail-over-IMAP a shot despite my distaste for Gmail's main Web interface.
In Mail.app I selected all 2100 messages in my Inbox and copied them over to Gmail. The operation completed without reported errors, but only 1300 messages made it over! Disturbed, I re-imported the original Inbox from a backup and copied it up to Gmail again, this time as a separate mailbox (label). Again, 1300 messages. As a control, I then copied that same re-imported mailbox up to my original account, and a few minutes later all 2100 messages were visible on the server. So it's not Mail.app.
Two bad things seem to be going on here. The first, and less severe, is that Gmail is running filters on these copied/moved messages as though they were new incoming mail - so some got flagged as spam or moved to other folders. While not a data loss issue, it's still annoying and not what one would expect.
The second is just pure data loss. Even after tallying all the filtered mail I came up about 500 messages short.
I'd love to think I'm missing something obvious here and there's a way around (since the extra, free storage would be lovely and even a mediocre AJAX UI is better than SquirrelMail), but clearly I can't trust a service that predictably loses messages.
Any thoughts?Well, I don't know how, but after removing all the accounts on my Pre (manually), manually backing up what remained to my Synergy/Palm Profile account, running WebOS Doctor once more and then re-adding all my accounts manually, Gmail is working again. I can't say exactly what fixed it, but man, what a pain!
Oh, and on a side note - nice going with the 2010 EAS (Exchange) calendar FAIL, Palm. I really appreciate that "little" bug too. *Rant off*
(http://forums.precentral.net/webos-synergy-synchronization/222080-calender-not-synching-new-year.htm...)
Message Edited by dpellenwood on 01-02-2010 06:19 PM -
Data loss -- managed photos mis-badged as missing referenced photos
I always import all photos as managed photos, and never use referenced photos. Yesterday, I noticed that the last 29 photos from a group of 96 photos imported Feb 18 were badged as "The referenced image’s master file has not been found."
Indeed, although I had worked on many of these images in Aperture shortly after importing them, I could now only view a large but low-res thumbnail.
Weirdly, when I right-click on one of these images and select "Manage Referenced Files..." I get a message that "The selection doesn't use any referenced files." I get the same message when I select "Manage Referenced Files..." from the File menu with the entire project selected.
Using the Finder, as well as the find and locate commands in the terminal, I have verified that the missing images are not on my main or backup hard drives (unfortunately, I use a simple rsync script for backup, and I didn't discover this problem until after the deletions had propagated to my backup).
The other 67 photos imported at the same time (same import group) seem fine, and, using the Finder, I can locate all of them in the Aperture library.
To test one user error scenario, I imported some test photos into the Aperture library and then relocated the masters to the desktop. I then quit Aperture, deleted the photos on the desktop, and relaunched Aperture. As expected, the test photos had the same "The referenced image’s master file has not been found" badge, but when I right clicked and selected "Manage Referenced Files..." I got the expected dialog, unlike when I select the same option for the missing 29 photos.
Any ideas what went wrong?
G5 iMac Mac OS X (10.4.8)Thanks for responding...
It sounds pretty odd. First try to rebuild the
database...as was suggested (Option>CMD on launch of
Aperture).
I will try this eventually. But I want to get some insight first, in case this could create more problems than it solves. I am also holding off on updating to 10.4.9 for the same reason.
This might help. If not, try to make an
adjustment on one of the images. This will force the
thumbnail to redraw. If the adjustments aren't
available for that particular image, then the image
is really offline somewhere.
I am unable to make any adjustments on these images.
I assume you went into your Library "Show Package
Contents" to confirm they are missing. If not, try
that.
Yes, the images are not in the folder with the other images from that import. Further, I used the unix 'find' and 'locate' commands to search my entire hard drive and backup drive, and the images were not found.
Can you Export the image?
That was how I first stumbled on this problem -- I tried to export and got an error message that "Versions with unavailable master images cannot be exported."
The weird thing is that when I try and get the 'Managed referenced files...' dialog, Aperture claims that "The selection doesn't use any referenced files" -- that tells me this data loss might be the result of a bug and not user error. -
Basically, any Pages file that has been modified and seemingly "overwritten" can be easily restored to its pre-5.0 state within Pages 5 by clicking File > Revert to > Browse All Versions and then selecting the previous version, clicking Restore and then closing Pages 5.0 without saving. That allows you to open the file in Pages '09 and delete Pages 5. Now everything is back to normal and you can continue using '09 until the feature set is restored.
I'm hoping people will continue to explore and verify this, but I've tried it myself and a few others have verified its validity. I started a thread yesterday but it quickly got buried in the swathes of complaints.
robogobo wrote:
Great! Hopefully this means everyone can get back to their old files and continue working in '09 until the feature set is restored to 5.0. Now we can all stop clamoring about Apple intentionally damaging users' work. I was 1000% sure that wasn't the case. After "Save As" was depricated, Autosave and Version History are part of a normal workflow (except for those who rejected the whole idea of auto save - but Apple didn't), which explains why files were not explicitly duplicated with the upgrade and no warning was issued.
Does this work for you, Peter? If so I hope you can see the value of actually using the product before jumping on the bashing bandwagon.
Karabuni wrote:
Seems to work - I've unplugged the external disk as well as turning off Time Machine. Old versions are there, and not only that, when the restored file is loaded into 5, it looks the same as it did in 4.3., whereas if just the latest one is opened, layout is, er, not right.
The "unrestored" file has layout errors, whereas, the "restored" one still looks OK when reopened, even after making a few changes. This might not be the case if a "removed" feature has been used, such as linked text boxes - they are there but not linked.
You will not be able to open this file in 4.3 if you make changes, as robogobo also said.
So, are the earlier versions inside the new one? Word docs have all sorts of old stuff inside if you open them in BBEdit, especially if the author has used an earlier letter to someone else as a template, and altered the text. Having looked at a 4.3 xml file there isn't any changed/deleted text or other obvious stored changes. So they must be somewhere else, I suppose.
The 5 files are zipped .iwa, and I haven't found anything that the system will allow to open them (unidentified developer).
robogobo wrote:
Could someone please verify this? Basically it would mean the transition from 4.3 to 5 is at least non destructive and we're dealing with a loss of features here but no data loss.
robogobo wrote:
Browse All Versions:
Now I hadn't tried this before, but it was on my todo list. When I first noticed problems with files opened in Pages 5.0, I was simply restoring files from Time Machine in order to get back to the original version for 4.3, thinking it was necessary for version history. But correct me if I'm wrong, Time Machine isn't necessary for browsing documents version history. Is that right? I just tested it and it seems to work with my Time Machine disk disconnected. Can someone verify this?
I just opened an old 4.3 document in 5.0, made some changes, saved and closed it. Then I tried to open it in 4.3 and of course it wouldnt open. So I opened it again in 5.0, went to File > Revert to > Browse All Versions and it opened up the interface, which looks like Time Machine (I always assumed it was) but I think it's not. Anyway, I went back to the old version, hit restore and then closed it without saving. I can now open it again in 4.3 in its original state!
Did everyone already know this? If it's true, then there isn't any permanent damage being done, and everything can be reverted to originals from within5.0. I think this is great news.Jeff Shenk wrote:
If you upgraded directly from Snow Leopard to Mavericks, and never had Versions in Pages 4, can you still revert to the oldest version in Pges 5 and get a Pages 4 document?
The documents brough into Mavericks from Snow Leopard should still be considered the original versions before they're opened in Pages 5, so I think it would work. Probably best to open them in 4.3 on Mavericks first though, to be safe.
At any rate, data loss is only a problem if you have opened the only copy of a document with a new, untried application, and in that case, you are acting as though it isn't a very important document, anyway.
Couldn't agree more. -
Data loss popup for cancel button on overview page
Hi Experts,
Data loss popup generally appears on UI, when i try to navigate with out saving the changes. This is a standard SAP behaviour.
But, i come across the requirement where this data loss popup should also appear on clicking the 'Cancel' button on overview page on UI.
I understood that, this popup will be triggered by the methods ON_BEFORE_WA_CONTENT_CHANGE & IF_BSP_WD_EVENT_HANDLER~HANDLE_EVENT of the window impl class.
Please suggest how can i make this popup work in case of Cancel button as well.
Thanks....Hi bkvs,
Usually when you click the "Cancel" button that means you don't want to save your entry... hence displaying the dataloss popup doesn't make really sense to me. However, you may want the user to confirm that he really wants to cancel.
So I would suggest to redefine the EH_ON_CANCEL method of your view, to display a simple confirmation popup before you actually cancel everything. This wiki will show you how to do it:
Creating popup message in webclient UI - CRM - SCN Wiki
Regards,
Nicolas. -
Updating to ios 6.1.5 leads to data loss in several apps (bento, teacher tool)
After updating to ios 6.1.5 I have experienced severe data loss in several apps. Can I return to older version of ios to rescue data?
All you can do is restore from backup
To restore from backup see:
iOS: How to back up
If you restore from iCloud backup the apps will be automatically downloaded. If you restore from iTunes backup the apps and music have to be in the iTunes library since synced media like apps and music are not included in the backup of the iOS device that iTunes makes.
You can redownload most iTunes purchases by:
Downloading past purchases from the App Store, iBookstore, and iTunes Store -
Friends,
I already had an iPad(1) and its full contents (not only the purchased items, applications from Apple, but my all own pics, videos, songs etc.) is normally syncronized with an authorized PC (Windows) computer.
Now, I've bought a new MacBook Air computer, and want to sell my PC. But I don't know how to transfer (or syncronize) all of the contents from my previous iTunes on PC to the new one on the Mac.
When I connected my iPad to my new MacBook Air, I promted, if I want to change authorization of my iPad to the MacBook.
But also got a warning, that the new iTunes will delete from my iPad all my previously syncronized items, pics, videos and music, with the previously authorized iTunes on the previous computer, expect the saved pics from the net.
What can I do just like that, simply to keep all my previous contents on my iPad, but have a new authorisation and a right syncronization, without data loss with the new iTunes on my new Mac?
Thanks a lot for your answer.
([email protected])Here are some instructions that I have posted several times that may be helpful to you as they have been to others. You can simply ignore anything that does not apply or that you have already done.
As Alan stated above - transferring the iTunes library is the best first step. If you can transfer the iTunes library to the Mac, most of this will be unecessary for you to do.
Very Important ....
1. Authorize the computer.
2. Turn off auto sync in iTunes
3. Transfer purchases
The text in italics is from the other thread - non italics are my words.
The following was copied from this thread. This is essentially what you want to accomplish.
https://discussions.apple.com/message/11527071#11527071
1) Without connecting your iPad to your new computer, start iTunes. Click on iTunes. Click on Preferences. Click on Devices. Check the box next to "Prevent your iPod etc. from automatically syncing." Click OK.
2) Now connect your iPad to your computer and start iTunes.
3) When iTunes starts, right click on your iPad under Devices in the left column. Select Transfer purchases etc.
4) After it finishes transferring all your apps to your computer, right click on your iPad and select Backup your iPad.
5) After it finishes backing up your iPad, right click on your iPad and select Restore etc.
6) After it finishes restoring, left click on your iPad , then click on the Apps tab on top, and check the box next to Sync Apps, then click on Apply below.
If everything on your iPad looks good after the sync, go back and click on iTunes / Preferences / Devices and UN-check the box next to Prevent your iPod etc. The only other thing you may want to check is if your contacts, bookmarks, etc. are syncing correctly now. If not, go to the Info tab after connecting and make sure you indicate which features you want to sync with what sources.
Read this thread and the support links as well. There are apps that you can purchase that will allow you to transfer photos from the iPad to your computer. Look at Photo Transfer App in the App store and you can search for others as well.
https://discussions.apple.com/message/13016026#13016026
This support site will help you with transferring iTunes music to your new computer.
http://support.apple.com/kb/HT4527
One final note - you may want to leave auto sync turned off - but that is totally up to you. I never did use auto sync.
Maybe you are looking for
-
LoadClass (error loading a class which extends other class at run-time)
Hey! I'm using the Reflection API I load a class called 'SubClass' which exists in a directory called 'subdir' at run-time from my program CustomClassLoader loader = new CustomClassLoader(); Class classRef = loader.loadClass("SubClass"); class Custom
-
My macbook pro (A1260) won't boot; powers up for 3 seconds and then shuts down.
Just recently my MBP A1260 gave up on me. When I push the power button, the dvd drive and hard drive both start, but the screen remains black. After 3 seconds, the hard drive makes a click and the whole laptop powers off. I suspect the hard drive has
-
ITunes not displaying Talking Walking podcasts – how can I resolve this?
iTunes not displaying Talking Walking podcasts – how can I resolve this? http://www.talkingwalking.net Created on iWeb version 3.0.4 Mac OS X 10.6.8 (10K549) Upgraded to Snow Leopard in August 2010 from Tiger. Originally published 26 podcast episodes
-
I have multiple users on my iMac. When I sign in as Admin, I give authorization in iCloud to "Find my Mac". When someone else logs in, the Find my Mac is grayed out and says Administrator authorization required. How do I fix this so no matter who is
-
If we want to transfer data to multiple receiver using context object, how
If we want to transfer data to multiple receiver using context object, how many receiver determinations we need to create?