[akent35]

I am not starting this thread to "start a war", so to speak. It just seems that almost every time I mention Super Duper as my backup solution, I get almost "bashed", per se, and a couple of individuals (no need to mention names) are adamant about it. In any event, let me go on.

I just did a google search of "Super Duper Vs. Time Machine", and one of the "hits" seems reasonable, accurate, and well put:

Straitmac Forum • View topic - SuperDuper vs Time Machine Backups

I completely understand how Time Machine can be useful for certain things, and Super Duper can also. One thing that is common to both programs (and Carbon Copy Cloner also) is that the "backups" (some folks use "cloning" for Super Duper and CCC) contain everything that is being backed up, even corrupt information. True, if Time Machine had made a prior backup that is not corrupt, all well and good. But, how does one know which backup is "non-corrupt"? Yes, at least with Time Machine, one can possibly find a non-corrupt backup. But, apparently it is not as easy to do a restore from a Time Machine backup as it is from a Super Duper or CCC "clone".

Also, Time Machine runs automatically, and is already free. With the other two, each one has to be purchased to use the automatic scheduling procedure (Super Duper can be used in "sponsored" mode, but there is no automatic scheduling).

Myself, for both of my machines, I do a weekly backup on my own. I actually first go through two other processes (Onyx and Tech Tool Pro (Disk Warrior can also be used)) to insure that what I am backing up is as "non-corrupt" as possible. Yes, it does take some time, but 1) the drives being looked at are smaller (256 gig SSDs), and 2) I can be doing other things while any of that software is running. For each machine, the entire process takes about 1 1/2 to 2 hours (I make two backups to external drives, and one of those is to a drive via a USB 2.0 connection, so it is slower).

Finally, for some strange reason, I recently got something corrupt that would only cause me problems while on this site (all other sites worked fine). Fortunately, that happened only two days after I had done my Super Duper Backup, so the Restore (a breeze, by the way) saved my "bacon", so to speak.

One other thing: I understand why one would need a very, very recent backup for their purposes, and yes, Time Machine is excellent for that (as apparently the paid versions of Super Duper and CCC are, regarding making frequent, timed backups). But, that is not my situation, and my processing works fine.

Again, none of this is meant to "start a war", and I hope folks will respond in a reasonable, beneficial way (apparently, according to the link I provided, that group has had some discussions also on this exact same topic).

[OldMacGeek]

No war necessary. My only thought is your use of the word "solution" - implying that backup is a singular event - is lacking. I tell people that the number of backups they have should be directly proportional to the importance of the data being backed up. While I'm a fan of Dave and his awesome software, I also use Time Machine for most-recent backups, and even ChronoSync for scheduled offsite backups. Of course, I'm paranoid. I also use rsync to mirror other datasets to other offsite locations, etc. I guess working for a storage company for a few decades can make you gun-shy. That and a wife whose journal articles, grant proposals, etc. are more precious than gold.

Again, I think you have a great solution that works for you. Believe me, you are so far ahead of the curve that you can sleep soundly. I know of many people who think that "backing up" means moving your data to an external hard drive and then deleting the original. Sigh . . .

[akent35]

Thanks for that perspective, OldMacGeek. These two statements by you are especially pertinent, and accurate:

"I tell people that the number of backups they have should be directly proportional to the importance of the data being backed up."

"That and a wife whose journal articles, grant proposals, etc. are more precious than gold."

That second one, of course, directly deals with your wife's job.

Again, thanks for the well stated reply.

[DesiSoftSystems]

Whatever product you use, it should back up and restore the full contents of your files and folders. I've experienced products that were developed by Windows-focused developers that prevented either backup or synchronization across OS X devices any file that contained characters illegal on Windows platforms but perfectly fine under OS X's default file system of HFS+. Another issue I've found is the belief that only the data fork of files should be saved or synchronized, an easy way to lose code-signing, Mavericks tags, and data relevant to an application. Also, to prevent any imagined issues with Windows users, the contents of bundled files (folders that appear as though a single file) are completely ignored by some of these products. Another issue is that files without extensions or that begin with a period character are ignored. Finally, while extreme, even though some file systems do not support files larger than 2 GB, or even larger than 4 GB, such files should also be backed up or synchronized.

I cannot name such products or their companies due to contracts in place and the possibility these issues will be rectified soon, but a way to test if a back up or synchronization product suffers from these symptoms, try backing up or synchronizing these:

1. In TextEdit, create a document with an embedded image in it, making sure to save it as a "Rich Text Document with Attachments".
2. In TextEdit, select a section of text and drag that text clipping to the Desktop.
3. In Mavericks or later, add a tag to a file.
4. In BBEdit, TextWrangler, or other text editing applications, change the character set of a text file that has the relevant sets of characters in it to an explicit type like UTF-8, UTF-16, MacKorean, etc.
5. Either in Finder or in the Save window of an application like TextEdit, name a file something like this:

<This folder has some unusual characters in it /=slash "=double quote |=vertical bar ?=question mark *=asterisk >

Note the spaces in the name should be included; if you like, you can make separate files that individually use any of the above unique, non-alphabetic characters.
6. If you are comfortable with Terminal, you could create a file named ".htaccess" with some data in it; this is used by web sites to manage the contents of a directory.
7. Many applications already add an extension automatically and may hide it, so doing that may be difficult, so you might have to use Terminal or Finder to remove an extension from an existing file.
8. The testing of the 2 GB or 4 GB limits are more difficult, but some apps may be distributed by way of DMG (Disk Image), and these are excellent test files.
9. In the Finder, using the Get Info window, change the icon of a folder to use a custom icon, possibly one copied from another file or folder.
10. In the Finder, using the Get Info window, change the icon of a file to use a custom icon, possibly one copied from another file or folder.

[pjosborne]

My apprach is to use both as I see both approaches as complimentary. What SuperDuper does is give me a fully bootable clone of my drive, such that if my Mac is stolen or the logic board fires I can *instantly* boot another Mac and have access to my applications and documets. I tend to view TimeMachine more as a version control system, allowing me to revert to an earlier version of a document or retrieve something I trashed by accident. Of course it can also fulfill the first role, but it requires a restore onto another HD, which is less than ideal if you’re borrowing someone else’s Mac and don’t have a spare external drive to hand.

[Charles Martin]

I agree with pjosborne, why not use both? I have a Time Machine backup drive, a second (bootable) backup made with Chronosync, a "media" drive and a cloned backup of that drive. Right now that means I have four hard drives next to my Mac, but I'm planning to consolidate this down a bit with larger-capacity drives in the future. There's also Carbon Copy Cloner, which is another excellent choice for those who want to make a bootable backup drive. I think bootable backups and Time Machine backups are complementary to each other as PJ said. For the really hard-to-replace data, like financial records or family photos etc, you'd also want to have an offsite backup (preferably an encrypted cloud backup) or those files on DVD or something like that.

[akent35]

Originally Posted by DesiSoftSystems 

Whatever product you use, it should back up and restore the full contents of your files and folders.

Super Duper does exactly that. As I mentioned in another forum, I recently needed to do a restore from my recent Super Duper backup on my Mac Mini. Worked like a charm, and everything was restored exactly as it was 2 days earlier, when I did the backup.

[akent35]

Originally Posted by chas_m 

I agree with pjosborne, why not use both?.

For my purposes, Super Duper is enough. As it is, I make 2 backup copies (on separate external devices). I feel rather safe with those.

Again, each one's requirements (and in many cases, employment needs) is different. Even when I was providing production support at home (a number of years ago), there was no need for me to make many backups of my data.

[mattyb]

I can't stress how important an off-site backup is. Think thief, water, tornado, fire.

[donaldkepler]

It is important to backup data, applications, files etc. with the reliable utilities. Because in case of something wrong, we are left only with the backup. There are many reliable utilities like Stellar Drive Clone, CCC for backing up data.

[amiller77]

I also recently started making a schedule bootable backup using CCC 1x/wk, which I then put in a fireproof strong box, in addition to Time Machine daily backups. I don't competely trust TIme Machine anymore because every several months, at least in the past year, it becomes corrupted somehow and refuses to backup anymore, necessitating a disk reformat of that external drive. In addition, using one of the available online backups such as DropBox (worked my way up to 6GB free by getting others to sign up) or Box.net (I snached up their recent offer of free 50GB of space) is a very smart move. You have to remember that once a file is deleted from the Cloud, it is deleted from all devices (although Dropbox at least saves recently deleted files for awhile). The files are encrypted, but I still don't upload financial records, just to be extra paranoid.

[akent35]

I don't use the cloud for anything, nor any other on line storage service. All my files, data, applications, backups, etc. are kept locally. No matter how many "promises" those cloud/on line services make, I will never trust them. It is just not worth the risk. Plus, for my needs, I don't need to access locally stored information while I'm away.

About the only 3 things I need access to while I'm away are 1) EMails, 2) Banking information, and 3) sometimes Stock Market quotas for our Mutual Funds. An internet connection solves that.

[davoud]

Originally Posted by akent35 

I don't use the cloud for anything, nor any other on line storage service...

I beg to differ. All of your household utilities are in the cloud. Your tax records, real-estate records, driver's license, and vehicle registration are in the cloud. If you shop with a credit or debit card your credit card and your shopping list are in the cloud. Your petrol purchases and the vegetables you buy from the supermarket are in the cloud.

If you use a smartphone with GPS or if you have an EZ-Pass (Northeast) or equivalent for highway tolls, your trips are in the cloud. If you have none of those you are in the cloud anyway because of speed cameras and cameras that read your license number in the cash lane at the toll booth. It will happen soon, if it hasn't already, that face-recognition software will have your identity and location in the cloud when you walk down the street.

Get over it. It's the way the world works today.

[davoud]

What's the big deal? I use SuperDuper and I use Time Machine. They work for me. Why should I care how, or whether, persons outside my household back up their data?

[andi*pandi]

Originally Posted by mattyb 

I can't stress how important an off-site backup is. Think thief, water, tornado, fire.

I know this is true, but every time I try to think about it, my mind just veers off. Like making a will. I know I should, but I haven't.

You just have to hear one horror story about someone losing their child's baby pictures, or in my case it would pretty much be all my photos for the last 12 years. Only some get Facebooked or Flickr'd.

[akent35]

Originally Posted by davoud 

I beg to differ. All of your household utilities are in the cloud. Your tax records, real-estate records, driver's license, and vehicle registration are in the cloud. If you shop with a credit or debit card your credit card and your shopping list are in the cloud. Your petrol purchases and the vegetables you buy from the supermarket are in the cloud.

If you use a smartphone with GPS or if you have an EZ-Pass (Northeast) or equivalent for highway tolls, your trips are in the cloud. If you have none of those you are in the cloud anyway because of speed cameras and cameras that read your license number in the cash lane at the toll booth. It will happen soon, if it hasn't already, that face-recognition software will have your identity and location in the cloud when you walk down the street.

Get over it. It's the way the world works today.

I guess I should have said that I do not, on my own, put stuff in the "cloud". But yes, all that other stuff you mentioned is in some kind of "cloud". It is the way of the world. But, I choose to control some of what one can put up there. That's my choice.

[akent35]

Originally Posted by andi*pandi 

I know this is true, but every time I try to think about it, my mind just veers off. Like making a will. I know I should, but I haven't.

You just have to hear one horror story about someone losing their child's baby pictures, or in my case it would pretty much be all my photos for the last 12 years. Only some get Facebooked or Flickr'd.

Yeah, andi, that is so, so true. I guess if I use the third external drive I have, a thin 1 TB 5400 rpm drive in a slim external case, backed up both of my machines to it, and stored it somewhere "secret", that could be a solution.

[turtle777]

Triple play for me:

1) Time Machine
2) Ext. HD backup that I store at work
3) All documents / pictures in the cloud (encrypted)

-t

[OreoCookie]

Originally Posted by andi*pandi 

I know this is true, but every time I try to think about it, my mind just veers off. Like making a will. I know I should, but I haven't.

Backblaze only costs $5/month. Their service is great, and I use it for my version of triple play:
(1) Time Machine
(2) Backblaze
(3) Dropbox

@akent35
I'm not exactly sure why you're posting here, but let me at least answer the two questions:

First of all, neither Time Machine nor SuperDuper tells you which data is not corrupt. But with Time Machine you can still recover older versions if you need to (and decide for yourself which one is not corrupt). Data integrity cannot be checked with most filesystems, only very modern ones like Oracle's ZFS or Linux's btrfs feature checksums on the data.

And it is not true that it is more difficult to recover from a Time Machine backup: all you need to do is boot the OS X Recovery partition, then launch the Restore From Time Machine Backup utility (it is one of the menu items). Select the backup you want to restore from (you do not have to pick the latest, you can restore to any backup). Then go out and grab a coffee. If the backup drive connects via USB or so, this is as fast as restoring from a SuperDuper backup because the bottleneck is the USB interface. Modern versions of OS X (10.7.3 and later) apparently even put a Recovery partition on your backup drive. That means in the worst case, you can boot also from your backup drive into recovery mode.

[akent35]

Originally Posted by OreoCookie 

@akent35
I'm not exactly sure why you're posting here, but let me at least answer the two questions:

First of all, neither Time Machine nor SuperDuper tells you which data is not corrupt. But with Time Machine you can still recover older versions if you need to (and decide for yourself which one is not corrupt). Data integrity cannot be checked with most filesystems, only very modern ones like Oracle's ZFS or Linux's btrfs feature checksums on the data.

Why can't I post here? I am the one that started this thread.

I never said that Time Machine or Super Duper can tell one which data is not corrupt. I'm not even sure where you got that from. What I do to minimize the corruption is to first run Oynx and Tech Tool Pro to help with this (I also occasionally run ClamXav to try and find any problematic data). It is after that that I run Super Duper to do my two backups. All of that is done with nothing else running (except the OS, of course). Additionally, between backups (I do it once a week), I am constantly cleaning things out, looking for (and apply) applicable updates, being extremely careful where I "surf", what I download, etc. Hence, the backups I make are as non-corrupt as possible, and thus I feel comfortable with restoring from such a backup. Not sure what else I can do to weed out as much corruption as I can.

If I were to use Time Machine for backups, I'm not sure how easy it is to control corruption. I would take the same precautions, but that means I would have to run Onyx and Tech Tool Pro (or Disk Warrior) much more often in order to minimize corrupt files/data.

In any event, I am comfortable with my strategy for minimizing corruption and backing up my information with Super Duper. Again, I am not saying it is a superior solution. Tieme Machine has quite a few good points, and for some folks, it is the way to go. I am fine using Super Duper.

Originally Posted by OreoCookie 

And it is not true that it is more difficult to recover from a Time Machine backup: all you need to do is boot the OS X Recovery partition, then launch the Restore From Time Machine Backup utility (it is one of the menu items). Select the backup you want to restore from (you do not have to pick the latest, you can restore to any backup). Then go out and grab a coffee. If the backup drive connects via USB or so, this is as fast as restoring from a SuperDuper backup because the bottleneck is the USB interface. Modern versions of OS X (10.7.3 and later) apparently even put a Recovery partition on your backup drive. That means in the worst case, you can boot also from your backup drive into recovery mode.

I had read on some other sites about how "difficult" it is to restore from a Time Machine backup, and thus I was just repeating what I had read. But, I can see where it is not that difficult, based on your explanation. A Super Duper restore is very easy to do: just boot from the Super Duper backup, start up Super Duper, select the source of the Restore and the destination, and let her "rip".

I also suspect the restores would take the same amount of time. I have never said (nor will I ever say) that one runs faster than the other.

[Spheric Harlot]

Originally Posted by akent35 

I never said that Time Machine or Super Duper can tell one which data is not corrupt. I'm not even sure where you got that from. What I do to minimize the corruption is to first run Oynx and Tech Tool Pro to help with this (I also occasionally run ClamXav to try and find any problematic data). It is after that that I run Super Duper to do my two backups. All of that is done with nothing else running (except the OS, of course). Additionally, between backups (I do it once a week), I am constantly cleaning things out, looking for (and apply) applicable updates, being extremely careful where I "surf", what I download, etc. Hence, the backups I make are as non-corrupt as possible, and thus I feel comfortable with restoring from such a backup. Not sure what else I can do to weed out as much corruption as I can.

None of those tools can tell you whether your data is corrupt, as they have nothing to check against. Onyx is entirely useless for that purpose, while TechTool Pro can only check for hardware defects. It has absolutely no way to tell whether, for example, Microsoft Word screwed up your file the last time you saved changes, or whether your iPhoto library has become corrupt due to a bug in the program that has deleted a bunch of originals that still show up as thumbnails.
And even if you repair any directory issues before cloning, there still is no recourse for data that has already been overwritten by overlapping files. You're merely cloning that messed-up file.

These cases are rare (except for the iPhoto thing; that happened all the time for a while a few years back), but guarding against those rare cases is, IMNSHO, the entire point of keeping a backup in the first place.

If I were to use Time Machine for backups, I'm not sure how easy it is to control corruption. I would take the same precautions, but that means I would have to run Onyx and Tech Tool Pro (or Disk Warrior) much more often in order to minimize corrupt files/data.

You could just stop wasting your time on that snake-oil and know that if you discover file corruption, you could just move backwards in time until the point where you find an intact version of the file.

[akent35]

Originally Posted by Spheric Harlot 

None of those tools can tell you whether your data is corrupt, as they have nothing to check against. Onyx is entirely useless for that purpose, while TechTool Pro can only check for hardware defects. It has absolutely no way to tell whether, for example, Microsoft Word screwed up your file the last time you saved changes, or whether your iPhoto library has become corrupt due to a bug in the program that has deleted a bunch of originals that still show up as thumbnails.
And even if you repair any directory issues before cloning, there still is no recourse for data that has already been overwritten by overlapping files. You're merely cloning that messed-up file.

These cases are rare (except for the iPhoto thing; that happened all the time for a while a few years back), but guarding against those rare cases is, IMNSHO, the entire point of keeping a backup in the first place.

Understand. But, using those tools is better than nothing. And, the point is that corruption would be present no matter which backup program one uses. And, for Time Machine backups (notice the plural), how does one know which backup to restore from? Yes, the option is there, but how does one somewhat easily get the right one?

The one tool that I mentioned which can help is ClamXav, and I guess I can run that more frequently (and in fact make it a part of my backup processing). Yes, it might not find everything that is "bad", but again it can help.

This discussion can keep going back and forth, but here are the essential points:

1. One needs to have a backup strategy, and one that works for them.

2. The choice of which "tools" to use for this is, again, what works for them.

3. Corruption of files can, at best, be minimized prior to doing a backup (whether done "automatically" like Time Machine, or manually like Super Duper). Again, one can use whatever tools are available to assist in that.

For my needs, I feel very comfortable with the processing I am using for this. It has been successful, I've never had an issue with it, and it works for me. For others, that is not sufficient, and they choose to use other tools. Fine. No one tool is superior to the other. Each has its' advantages and disadvantages.

Originally Posted by Spheric Harlot 

You could just stop wasting your time on that snake-oil and know that if you discover file corruption, you could just move backwards in time until the point where you find an intact version of the file.

That snake oil also contains no venom, so doing nothing can introduce snake venom into one's files. If you want that, fine. I'll continue to keep as much venom out as possible.

[OreoCookie]

@akent35
If you want to read a very recent overview article of backup strategies, I recommend this gem on Anandtech.com. It gives an overview of built-in and third-party backup software, cloud services and how to arrive at a backup strategy tailored to your needs.

Originally Posted by akent35 

I never said that Time Machine or Super Duper can tell one which data is not corrupt. I'm not even sure where you got that from. What I do to minimize the corruption is to first run Oynx and Tech Tool Pro to help with this (I also occasionally run ClamXav to try and find any problematic data).

TechTool Pro (and similar tools) only check for filesystem integrity -- which is important. They cannot check for things like bit rot and such -- unless you manually checksum your files and compare. So basically since consumer-level file systems do not support check sums for data, we are not protected.

Originally Posted by akent35 

If I were to use Time Machine for backups, I'm not sure how easy it is to control corruption. I would take the same precautions, but that means I would have to run Onyx and Tech Tool Pro (or Disk Warrior) much more often in order to minimize corrupt files/data.

Well, you don't control corruption, but you have the chance that you still have an uncorrupted copy of your file in your Time Machine backup. With clones, all bets are off once you have cloned corrupt data. Clones of your hard drive can help you minimize downtime in case of a hard disk failure (you can work off the clone, for instance). But clones by themselves are a very bad backup solution, because you don't have access to older versions of your files. In contrast, Time Machine automatically makes hourly backups. Not having to rely on user interaction is a huge plus for any backup solution.

Originally Posted by akent35 

I had read on some other sites about how "difficult" it is to restore from a Time Machine backup, and thus I was just repeating what I had read. But, I can see where it is not that difficult, based on your explanation. A Super Duper restore is very easy to do: just boot from the Super Duper backup, start up Super Duper, select the source of the Restore and the destination, and let her "rip".

You can also use Disk Utility to clone volumes -- and that's also as simple as restoring from Time Machine or using SuperDuper. I've used Time Machine to migrate from machine to machine for several years and never had a problem nor found it difficult.

Originally Posted by akent35 

I also suspect the restores would take the same amount of time. I have never said (nor will I ever say) that one runs faster than the other.

Not necessarily: the fastest way is to use Disk Utility to clone one volume from another (or an iso file), because here, the copies are done on the bit level. Time Machine works on the file level so it can be slower. However, that doesn't matter unless you connect your external drive via SATA or Thunderbolt, because then the interface is the limiting factor.

[akent35]

Originally Posted by OreoCookie 

@akent35
If you want to read a very recent overview article of backup strategies, I recommend this gem on Anandtech.com. It gives an overview of built-in and third-party backup software, cloud services and how to arrive at a backup strategy tailored to your needs.

While a good article, it did not show any third party alternatives (at least on a Mac, as I skipped the stuff on Windows). At least the link I initially posted did a fair comparison between built-in Time Machine and third party Super Duper (a few other third party ones were mentioned also).

Originally Posted by OreoCookie 

TechTool Pro (and similar tools) only check for filesystem integrity -- which is important. They cannot check for things like bit rot and such -- unless you manually checksum your files and compare. So basically since consumer-level file systems do not support check sums for data, we are not protected.

Understand. And, that is valid no matter which backup software/method one uses.

Originally Posted by OreoCookie 

Well, you don't control corruption, but you have the chance that you still have an uncorrupted copy of your file in your Time Machine backup. With clones, all bets are off once you have cloned corrupt data. Clones of your hard drive can help you minimize downtime in case of a hard disk failure (you can work off the clone, for instance). But clones by themselves are a very bad backup solution, because you don't have access to older versions of your files. In contrast, Time Machine automatically makes hourly backups. Not having to rely on user interaction is a huge plus for any backup solution.

Your use of the word "bad" is not accurate. In fact, I'd say just the opposite: very good (note that I did not say superior. Need to keep things fair and balanced.). There are enough positive reviews of products like Super Duper and Carbon Copy Cloner that show they are good. Also, the strategies one uses with them is also relevant. I feel my strategy is fine, sufficient, and good enough for my needs.

By the way, when Time Machine does its' hourly backups, I'm assuming that is not necessarily done "in the background", so to speak. That is, one can also be doing other things while Time Machine is running. If that is the case, at what point is the backup being done? If it "froze" the system (for its' backup purposes) at the time it started, that's ideal. When I start my backup processing, except for the OS, I have nothing else running. That is, I run Oynx by itself, then Tech Tool Pro by itself, and then Super Duper by itself. I thus have a "clean" backup, per se, and done at a very specific, and stable, point in time. I prefer that. But, I understand if others want to use Time Machine.

Originally Posted by OreoCookie 

You can also use Disk Utility to clone volumes -- and that's also as simple as restoring from Time Machine or using SuperDuper. I've used Time Machine to migrate from machine to machine for several years and never had a problem nor found it difficult.

Understand. And, I've used Super Duper for so long, I depend upon it (I recently had to do a restore, and it worked like a charm!).

Originally Posted by OreoCookie 

Not necessarily: the fastest way is to use Disk Utility to clone one volume from another (or an iso file), because here, the copies are done on the bit level. Time Machine works on the file level so it can be slower. However, that doesn't matter unless you connect your external drive via SATA or Thunderbolt, because then the interface is the limiting factor.

Understand that, too. When I do one of my backups on my Mac Mini, I use the external device's Firewire 800 port (has two of them, and the Mac Mini has one). Definitely faster than the other backup on the other external drive, which is via USB 2.0. For my Mac Book Air, and that first external hard drive, I use a Firewire 800 - to - Thunderbolt adapter, as the Mac Book Air only has a USB 3.0 port, and a Thunderbolt port. Still faster than with the other backup (via USB 2.0).

[reader50]

Originally Posted by akent35 

By the way, when Time Machine does its' hourly backups, I'm assuming that is not necessarily done "in the background", so to speak. That is, one can also be doing other things while Time Machine is running. If that is the case, at what point is the backup being done? If it "froze" the system (for its' backup purposes) at the time it started, that's ideal. When I start my backup processing, except for the OS, I have nothing else running. That is, I run Oynx by itself, then Tech Tool Pro by itself, and then Super Duper by itself. I thus have a "clean" backup, per se, and done at a very specific, and stable, point in time.

Time Machine does a follow-up pass before completion, covering only files changed since the backup started. You can see this in Console - select All Messages, then type "backupd" in the search box. I think I've seen a still-smaller 3rd pass sometimes, if there were file changes during the 2nd pass.

Not sure how open files are handled. TM might record a snapshot anyway. Or skip all files that are open during the entire backup process. The resulting TM backup would show the most recent copies of such files, made when each was last closed during a backup.

[akent35]

Originally Posted by reader50 

Time Machine does a follow-up pass before completion, covering only files changed since the backup started. You can see this in Console - select All Messages, then type "backupd" in the search box. I think I've seen a still-smaller 3rd pass sometimes, if there were file changes during the 2nd pass.

Not sure how open files are handled. TM might record a snapshot anyway. Or skip all files that are open during the entire backup process. The resulting TM backup would show the most recent copies of such files, made when each was last closed during a backup.

Thanks. I was wondering how that happened.

So, if a file has changed (and becomes corrupt, for any reason), that one gets backed up (as long as another change to the file does not occur prior to Time Machine completing its' job). To me, that seems risky. I would rather do the backup while things are static and stable. Since I do it after my primary tasks are completed on Saturdays, I'm good to go.

[reader50]

While TM is doing a backup, it will keep recording latest versions of changed files, until no file changes happen during a pass. This becomes the completed snapshot. Note, this refers to known file changes, ie: write operations. Or file delete, truncate, etc. Bitrot in place would not be recorded until the next time that file is modified a normal way.

The idea is regular background backups, for all those who otherwise forget or put it off. I expect the advantages outweigh the disadvantages for the vast majority of users.

The bitrot problem is going to get worse as our filespaces keep growing. We really do need those file checksums, and filesystem redundancy to correct. This still won't protect against an app writing out a corrupt file.

[OreoCookie]

Originally Posted by akent35 

While a good article, it did not show any third party alternatives (at least on a Mac, as I skipped the stuff on Windows). At least the link I initially posted did a fair comparison between built-in Time Machine and third party Super Duper (a few other third party ones were mentioned also).

They do mention quite a few cloud services such as Backblaze or Amazon S3. The latter works with many third-party backup tools for the Mac, e. g. xTwin.

Originally Posted by akent35 

Understand. And, that is valid no matter which backup software/method one uses.

Not really. Some backup strategies span a larger safety net than others. And with clones you just have 1 additional copy. If you have backups of older versions, the chance that you may recover something is much larger.

Originally Posted by akent35 

There are enough positive reviews of products like Super Duper and Carbon Copy Cloner that show they are good. Also, the strategies one uses with them is also relevant.

SuperDuper and CCC are good tools for what they do, cloning, no doubt about that. But their use as a backup tool is outdated. This is what the developer of SuperDuper has to say about that (emphasis mine):

Originally Posted by shirt-pocket.com

Time Machine
As far as Time Machine goes -- in general, I've always thought that some sort of backup functionality belongs in the OS. It's been a long time coming. The fact that it wasn't there left opportunities for 3rd parties, but that doesn't mean Apple shouldn't address the missing functionality.

And so, they have, with Time Machine. Really, I think that's a great thing. People need to back up more often, and I think Time Machine encourages them to do so, and gives them a relatively transparent way to do it.

Time Machine and SuperDuper!
That said, though, Time Machine isn't the be-all and end-all of backup programs. In fact, given how it works, I really do think that SuperDuper! remains both relevant and necessary -- a true complement to the functionality in Time Machine.

First, as is likely obvious, Time Machine is designed to provide automatic "temporal" backup (discussed in broad terms in the post The Ninety-Nine-Per-Cent Solution many months ago). Its primary usage scenario -- and the one that most demonstrations and documentation focuses on -- is to allow quick recovery of files and data that have gone missing, etc.

It does this in a way that's highly integrated with the OS, with a unique UI that's both cool and kinda cheesy (I'll tell you, the 'space' theme hasn't grown on me at all...)... and, as was the case with Spotlight, with a certain amount of application-level impact (something 3rd parties like Shirt Pocket could never mandate).

Fully, Directly Bootable Backups with SuperDuper!
What's important to note is that this isn't, and never was, what SuperDuper! was designed to do.

Our tagline, Heroic System Recovery for Mere Mortals, tries to sum up the whole idea: SuperDuper! is designed to [n]provide excellent failover support for the all-too-common case where things fail in a pretty catastrophic way, such as when a drive fails, or your system becomes unbootable.[/b] We do this by quickly and efficiently creating a fully bootable copy of your source drive. Perhaps more importantly, recovery is near immediate, even if the original drive is completely unusable, because you can start up from your backup and continue working.

You can even take your backup to a totally different Macintosh, start up from it, and work while your failed Macintosh is in the shop... then, when it comes back all fresh and shiny, restore things and keep working. And even if the other Mac is a different CPU type, you can still open and edit the files on the backup.

You cannot do this with Time Machine: Time Machine copies are not bootable until they're restored.

So even the developer envisions that his product be used alongside a tool like TimeMachine -- and I agree with the developer, it's a great tool for the use case he has in mind. And SuperDuper can, of course, be part of a backup strategy -- as long as it is not the only piece of the puzzle. By the way, also the OP of the thread you link to in your first post uses Time Machine and SuperDuper side-by-side.

Originally Posted by akent35 

By the way, when Time Machine does its' hourly backups, I'm assuming that is not necessarily done "in the background", so to speak.

It's done in the background, and it does not freeze the system. OS X allows only one process to access a file so OS X will not backup (= read) a file while another application writes to it. The system is not frozen, you can continue to work on it. Typically, the amount of data that needs to be shoveled back and forth is very small, because Time Machine keeps book on which files have been changed since the last backup (unless you reboot, then its memory is wiped).

Originally Posted by akent35 

So, if a file has changed (and becomes corrupt, for any reason), that one gets backed up (as long as another change to the file does not occur prior to Time Machine completing its' job). To me, that seems risky.

That can't happen by design: only one process can update the filesystem at any time. To be honest, nowadays such a giant lock is archaic, but it ensures that the situation you have in mind just can't happen on a Mac, period. I don't know how opened files are handled either, though, maybe they are skipped.

Originally Posted by akent35 

When I start my backup processing, except for the OS, I have nothing else running.

You have plenty of things running, you have tens of processes that could, e. g. download and install updates in the background. Modern backup software is not dependent on it being the »only« thing that's running on the system.

[OreoCookie]

Originally Posted by reader50 

The bitrot problem is going to get worse as our filespaces keep growing. We really do need those file checksums, and filesystem redundancy to correct. This still won't protect against an app writing out a corrupt file.

That's worth pointing out, it's just a matter of expectation value: even if, say, the probability of a bit flip within the time span of a year is 1 : 1 billion = 1 : 10^9, you end up with thousands of flipped bits on a 1 TB. We really, really need file checksums in file systems. (I've tried looking up real probabilities, but I was unable to find anything.)

[P]

Originally Posted by OreoCookie 

That's worth pointing out, it's just a matter of expectation value: even if, say, the probability of a bit flip within the time span of a year is 1 : 1 billion = 1 : 10^9, you end up with thousands of flipped bits on a 1 TB. We really, really need file checksums in file systems. (I've tried looking up real probabilities, but I was unable to find anything.)

We need some sort of checksumming, but there is one point that is often overlooked when discussing ZFS on a Mac: ZFS was designed with the assumption that all computers it would ever run on used ECC RAM. The only Mac that does is the Mac Pro, a vanishingly small percentage of all Macs in use. The same math that you did above applies to RAM as well, and the target area there has grown with the RAM capacity. My (almost) five year old Mac has 16 GB RAM, and it is just as likely to have a soft flip as that 1MB Mac SE I had in the eighties - as likely per bit of RAM, that is, so in total about 16384 times as likely.

The trouble of course is Intel, which charges through the nose for its server chips and disables ECC support in its desktop chips to sell them.

[P]

Originally Posted by akent35 

Understand. And, that is valid no matter which backup software/method one uses.

No, it is not. If you try to use a file that you haven't used in a long time and discover that it has been corrupted, Time Machine has your back and SuperDuper, generally, does not. Using Time Machine, you can restore that file from one week ago and see if it is fine, and if not restore it from a month ago, two months ago, a year ago or back to your first backup. The only chance that Super Duper will help you is that you have saved a clone from before the corruption. You will need a lot of HDs if you plan to save anywhere near as many backups as Time Machine does.

This is the part we always get back to. Nobody is disputing that SuperDuper is a great program or that it has advantages over Time Machine backups in some cases, but in this case, Time Machine helps you and SuperDuper does not. That you have had a good experience with your strategy so far is only because you still haven't had a file corruption problem yet.

[Spheric Harlot]

Indeed.

Cloning is a useful tool if downtime is expensive and it is vital to get a system up and running again in a known state immediately. Its use is in creating a SYSTEM backup.

It is NOT a DATA backup strategy.

[OreoCookie]

Originally Posted by P 

We need some sort of checksumming, but there is one point that is often overlooked when discussing ZFS on a Mac: ZFS was designed with the assumption that all computers it would ever run on used ECC RAM.

You're absolutely right, but it's even more general than that: ZFS was designed to run on a server with lots of resources where in particular power is also of no objective. So certainly, ZFS (or any other copy-on-write filesystem) is not a good fit for Macs (most of which are mobile) and iOS devices (yes, I want file safety also on my iPhone, thank you very much).

To me, one potential solution is a combined software/hardware approach. The SSD controller and parts of the filesystem are doing the same job on top of one another (e. g. SSDs do copy-on-write anyway but standard file system do not know about this and the SSD controller is managing the blocks). I reckon you can implement checksumming (and perhaps some other file system features we lust) for in the SSD controller -- in hardware if possible, in firmware if necessary. I still think it'll be a hard problem to solve, but it's a problem we need to solve.

[P]

Many SSDs also do deduping in hardware, but that is invisible to the operating system, so it looks as if more space is in use than is actually the case. Our old storage model is creaking, it was never built to handle multi-TB drives that have internal logic. ZFS is a way to try cut the Gordian knot by integrating the LVM into the FS, but the SATA command set limits what it can do. We would need to rethink the entire thing all the way down to the flash silicon to figure this out, and that seems more likely to happen in mobile than on the desktop.

Anyway: there is actually some good news on the horizon. Samsung recently announced flash with multiple cells stacked vertically, and since this means that the density skyrockets, they can make them on older larger processes, which improves reliability. The density increase is big enough anyway that we might just be able to ditch the old spinning rust completely in a few years, which would be fantastic for reliability.

Even better, DDR4 is finally getting close, as Haswell-E, launching this quarter, will support it. Even if ECC is not mandated (although Haswell-EP will have it, it is not required by the standard), there is some new error correction - the signals between CPU and RAM are now error-corrected, removing one more point of failure. We still have the original soft flips - cosmic radiation isn't going away - but that problem may not be quite so large as it seemed a few years ago.

[ghporter]

Originally Posted by Spheric Harlot 

Indeed.

Cloning is a useful tool if downtime is expensive and it is vital to get a system up and running again in a known state immediately. Its use is in creating a SYSTEM backup.

It is NOT a DATA backup strategy.

I think this is an incredibly important point. A fully installed system is a great target to clone; it gives you a "clean slate" machine state, and has (with Windows) saved my cookies many times. But while a fully configured yet pristine system installation should be pretty much static, everyone's data is often more fluid than they even can be aware of. And since it can change so often and so subtly, it is at risk for data corruption and data loss. Thus an error-checked, versioned backup of one's data is important. Cloning makes as exact a copy of what's being cloned, errors and all, while a backup app like Time Machine checks filles for errors while it's working.

[OreoCookie]

Originally Posted by P 

Many SSDs also do deduping in hardware, but that is invisible to the operating system, so it looks as if more space is in use than is actually the case.

Nice. And many also do on-the-fly compression.

Originally Posted by P 

Anyway: there is actually some good news on the horizon. Samsung recently announced flash with multiple cells stacked vertically, and since this means that the density skyrockets, they can make them on older larger processes, which improves reliability. The density increase is big enough anyway that we might just be able to ditch the old spinning rust completely in a few years, which would be fantastic for reliability.

That'd be great. This year was the first time ever, my »hard drive« size decreased when I switched to a Retina MacBook Pro.

[akent35]

Originally Posted by OreoCookie 

They do mention quite a few cloud services such as Backblaze or Amazon S3. The latter works with many third-party backup tools for the Mac, e. g. xTwin.

The article I provided the link to discusses both Time Machine and Super Duper equally. The one you provided deals just about exclusively with built-in backup software.

Originally Posted by OreoCookie 

Not really. Some backup strategies span a larger safety net than others. And with clones you just have 1 additional copy. If you have backups of older versions, the chance that you may recover something is much larger.

I have said before that I understand. I have also said that I firmly believe that I take just about all the precautions to minimize that.

Originally Posted by OreoCookie 

SuperDuper and CCC are good tools for what they do, cloning, no doubt about that. But their use as a backup tool is outdated. This is what the developer of SuperDuper has to say about that (emphasis mine):

So even the developer envisions that his product be used alongside a tool like TimeMachine -- and I agree with the developer, it's a great tool for the use case he has in mind. And SuperDuper can, of course, be part of a backup strategy -- as long as it is not the only piece of the puzzle. By the way, also the OP of the thread you link to in your first post uses Time Machine and SuperDuper side-by-side.

Tools such as Super Duper and Carbon Copy Cloner are definitely not outdated. As it is, you "conveniently" omitted stating an obvious (and serious defect with Time Machine). I quote:

"You cannot do this with Time Machine: Time Machine copies are not bootable until they're restored."

Super Duper (and Carbon Copy Cloner) can do a full restore after its' bootable "image" is started up. That, at least for me, is a BIG plus (and it recently was just that for me).

Originally Posted by OreoCookie 

It's done in the background, and it does not freeze the system. OS X allows only one process to access a file so OS X will not backup (= read) a file while another application writes to it. The system is not frozen, you can continue to work on it. Typically, the amount of data that needs to be shoveled back and forth is very small, because Time Machine keeps book on which files have been changed since the last backup (unless you reboot, then its memory is wiped).

You left off what I stated immediately after. To quote:

"By the way, when Time Machine does its' hourly backups, I'm assuming that is not necessarily done "in the background", so to speak. (Bold for emphasis, showimng what you forgot) That is, one can also be doing other things while Time Machine is running. If that is the case, at what point is the backup being done? If it "froze" the system (for its' backup purposes) at the time it started, that's ideal."

Someone subsequently clarified that for me.

Originally Posted by OreoCookie 

You have plenty of things running, you have tens of processes that could, e. g. download and install updates in the background. Modern backup software is not dependent on it being the »only« thing that's running on the system.

I have never allowed automatic downloads of updates, and subsequent installations. I am just about certain I have that shut off, especially within Mavericks itself. If I am missing something by doing that, then it would be good to know. But, I have no other applications running at all when I start my backup processing. (I guess if this is an issue, I could disconnect my Telephony modem from my machine).

[akent35]

Originally Posted by P 

No, it is not. If you try to use a file that you haven't used in a long time and discover that it has been corrupted, Time Machine has your back and SuperDuper, generally, does not. Using Time Machine, you can restore that file from one week ago and see if it is fine, and if not restore it from a month ago, two months ago, a year ago or back to your first backup. The only chance that Super Duper will help you is that you have saved a clone from before the corruption. You will need a lot of HDs if you plan to save anywhere near as many backups as Time Machine does.

Understand. But again, I believe I am taking all the "necessary as possible" precautions during the entire time between my weekly backups.

Originally Posted by P 

This is the part we always get back to. Nobody is disputing that SuperDuper is a great program or that it has advantages over Time Machine backups in some cases, but in this case, Time Machine helps you and SuperDuper does not. That you have had a good experience with your strategy so far is only because you still haven't had a file corruption problem yet.

The biggest advantage that Super Duper (and Carbon Copy Cloner) has, and most serious defect of Time Machine, is what was mentioned above. That is, "You cannot do this (restore from a bootable backup) with Time Machine: Time Machine copies are not bootable until they're restored." For me, that is priceless!

[akent35]

Originally Posted by OreoCookie 

This year was the first time ever, my »hard drive« size decreased when I switched to a Retina MacBook Pro.

Same for me, but last year. I purchased my MacMini last July, and it had a slow 5400 rpm 1 TB drive. Three months later, I replaced it with a fast, 256 gig Samsung SSD (the 1 TB drive went into a nice, slim external case). And, when I purchased my 13" Mac Book Air last Columbus Day (got it for a great price!), it had a 256 gig SSD.

In both cases, neither drive is even half full, but I like having the extra free space, and that does, at times, come in handy.

[P]

Originally Posted by akent35 

Understand. But again, I believe I am taking all the "necessary as possible" precautions during the entire time between my weekly backups.

So you're what, checksumming every single file and comparing to the last backup, restoring any file where the checksum changed and the modification date didn't?

The biggest advantage that Super Duper (and Carbon Copy Cloner) has, and most serious defect of Time Machine, is what was mentioned above. That is, "You cannot do this (restore from a bootable backup) with Time Machine: Time Machine copies are not bootable until they're restored." For me, that is priceless!

Poor man's RAID, essentially. That is a relevant feature, if those are your needs. But you still haven't showed anything to explain how you handle file corruption.

[Spheric Harlot]

Originally Posted by akent35 

The biggest advantage that Super Duper (and Carbon Copy Cloner) has, and most serious defect of Time Machine, is what was mentioned above. That is, "You cannot do this (restore from a bootable backup) with Time Machine: Time Machine copies are not bootable until they're restored." For me, that is priceless!

And that's fine, because keeping an immediately bootable system duplicate is exactly what the tool is there for.

It is NOT, however, a backup. It is a duplicate of whatever was there, in whatever state it was, intact or not.

[P]

Originally Posted by OreoCookie 

That'd be great. This year was the first time ever, my »hard drive« size decreased when I switched to a Retina MacBook Pro.

More about Samsung's 3D flash here. 32 layers at 40nm, and Sammy claims that they will double capacity every year until 2017 - and they already sell a 1TB version.

Note that the Pro in the name means that it is the enterprise stuff. No consumer drives launched yet, AFAIK, and the enterprise stuff is usually quite expensive. In the long run, this should not be expensive. 40 nm sheets are cheap by now, and the defect rate should be tiny.

[akent35]

Originally Posted by P 

So you're what, checksumming every single file and comparing to the last backup, restoring any file where the checksum changed and the modification date didn't?

Man, I already stated above what I try and do. In fact, I have no idea about "checksum", nor do I have software that can do that.

Originally Posted by P 

Poor man's RAID, essentially. That is a relevant feature, if those are your needs. But you still haven't showed anything to explain how you handle file corruption.

Might seem like a "Poor man's RAID", but it works for me, and with a product that has numerous positive reviews.

See above regarding file corruption.

And, you still have not explained that serious defect in Time Machine that I mentioned: cannot boot from it.

[akent35]

Originally Posted by Spheric Harlot 

And that's fine, because keeping an immediately bootable system duplicate is exactly what the tool is there for.

It is NOT, however, a backup. It is a duplicate of whatever was there, in whatever state it was, intact or not.

Yes, it is a duplicate of whatever was there. Same goes for Time Machine backups. Corrupt or not, it makes a backup.

As I originally stated, some folks on here call using products like Super Duper and Carbon Copy Cloner "cloning", but to me, that is the same as a backup.

Here is a quote from one of the links I got from a google search of "backup":

"In Information Technology, a backup, or the process of backing up, refers to the copying and archiving of computer data so it may be used to restore the original after a data loss event."

Similarly, here is a quote from one of the links I got from a google search of "cloning in Information Technology":

"Disk cloning is the act of copying the contents of a computer's hard drive. The contents are typically saved as a disk image file and transferred to a storage medium, which could be another computer's hard drive or removable media such as a DVD or a USB drive."

The word "copying" is used in both of those. And, "cloning" does back up "date" files.

[prl99]

Originally Posted by davoud 

I beg to differ. All of your household utilities are in the cloud. Your tax records, real-estate records, driver's license, and vehicle registration are in the cloud. If you shop with a credit or debit card your credit card and your shopping list are in the cloud. Your petrol purchases and the vegetables you buy from the supermarket are in the cloud.

If you use a smartphone with GPS or if you have an EZ-Pass (Northeast) or equivalent for highway tolls, your trips are in the cloud. If you have none of those you are in the cloud anyway because of speed cameras and cameras that read your license number in the cash lane at the toll booth. It will happen soon, if it hasn't already, that face-recognition software will have your identity and location in the cloud when you walk down the street.

Get over it. It's the way the world works today.

You're stretching the concept of the "cloud." My tax records are contained on IRS servers (which better not be hosted on Amazon cloud services), real estate records on the title company's and assessor's servers, driver's license on the DMV's servers, and vehicle registration on the DOL servers. These are accessible from the internet but aren't necessarily hosted on cloud servers. There's a big difference. You other examples are also housed on specific servers housed either in the local government or corporate computer center or split between several servers at sites around the nation. This doesn't mean they're in the cloud. The term cloud is now used for all sorts of storage locations that were never termed "in the cloud" before. Even iCloud is still a collection of server farms owned by Apple for use by Apple users. Amazon and IBM cloud services are different. They are server farms created by a company that in turn sells storage services to whomever wants to pay for them for both specific and general usage.

[akent35]

Originally Posted by prl99 

You're stretching the concept of the "cloud." My tax records are contained on IRS servers (which better not be hosted on Amazon cloud services), real estate records on the title company's and assessor's servers, driver's license on the DMV's servers, and vehicle registration on the DOL servers. These are accessible from the internet but aren't necessarily hosted on cloud servers. There's a big difference. You other examples are also housed on specific servers housed either in the local government or corporate computer center or split between several servers at sites around the nation. This doesn't mean they're in the cloud. The term cloud is now used for all sorts of storage locations that were never termed "in the cloud" before. Even iCloud is still a collection of server farms owned by Apple for use by Apple users. Amazon and IBM cloud services are different. They are server farms created by a company that in turn sells storage services to whomever wants to pay for them for both specific and general usage.

Well, well stated, prl99! The "cloud services" that one chooses to use individually are clearly different from the ones you mentioned in your second through fifth sentences.

Again, thanks for that excellent clarification!

[OreoCookie]

Originally Posted by akent35 

The article I provided the link to discusses both Time Machine and Super Duper equally. The one you provided deals just about exclusively with built-in backup software.

The anandtech article dedicates several sections/pages to third-party backup options such as cloud services and RAIDs.

Originally Posted by akent35 

Tools such as Super Duper and Carbon Copy Cloner are definitely not outdated. As it is, you "conveniently" omitted stating an obvious (and serious defect with Time Machine). I quote:

I haven't omitted that at all, for instance, I wrote that »Clones of your hard drive can help you minimize downtime in case of a hard disk failure (you can work off the clone, for instance).« Also I put emphasis on this when I quoted the FAQ from the developer of SuperDuper (last emphasis to be precise). If you intend to use the cloned drive to cut down your down time in case your primary drives no longer works, then that's exactly what clones are for.

Time Machine is not flawless (I'm particularly worried about HFS+ corruption and bit rot), but it's not the only backup tool I use -- and it certainly is a lot, lot, lot better than anything that existed before. And it's free. It's not a defect of Time Machine that you can't boot off your Time Machine drive, because that's not what Time Machine was designed for -- Time Machine was designed to be an easy to set up incremental backup, and not a cloning software.

Originally Posted by akent35 

I have never allowed automatic downloads of updates, and subsequent installations. I am just about certain I have that shut off, especially within Mavericks itself. If I am missing something by doing that, then it would be good to know. But, I have no other applications running at all when I start my backup processing. (I guess if this is an issue, I could disconnect my Telephony modem from my machine).

Launch Activity Monitor and look at all the processes which are running even when you have no apps running: it's a lot (make sure to show all processes). With modern backup tools such precautions are no longer necessary because they have been designed for »modern« multitasking operating systems (I write modern in quotation marks, because it was modern 20 years ago). There are backup tools for databases, for instance, that can be used while the database is running.

Look, if you just came here to defend your way of doing backups with SuperDuper, then let's stop here. But it should be a clear sign to you that not just all of us here who participated in this thread, but also Anandtech, SuperDuper's software developer and the guy whose post you linked to also use and recommend Time Machine -- possibly in addition to cloning software, but not the other way around.

[OreoCookie]

Originally Posted by prl99 

The term cloud is now used for all sorts of storage locations that were never termed "in the cloud" before. Even iCloud is still a collection of server farms owned by Apple for use by Apple users. Amazon and IBM cloud services are different. They are server farms created by a company that in turn sells storage services to whomever wants to pay for them for both specific and general usage.

Indeed, cloud is a very vague terms and some people are afraid of it -- even though they've been using something that people would now claim is »in the cloud«. There are ways to protect your data or take complete ownership of it. You can create your own »cloud« with open source tools on your own server. You can use Transporters and you will always keep control of the data (I have one at home: if I disconnect it, nobody can access the data on my Transporter. It is quite reasonable that some people are not willing to put certain data on their Dropbox, for instance. In fact, privacy laws in certain jurisdictions may forbid this explicitly.

But there are backup solutions which protect your data in those instances: xTwin, for instance, encrypts all traffic on the client before it sends anything out to, say, Amazon S3.

[akent35]

Originally Posted by OreoCookie 

The anandtech article dedicates several sections/pages to third-party backup options such as cloud services and RAIDs.

My first post of this thread was mainly geared towards desktop software for backups, and the link I provided was an example of that. I already stated why I don't like cloud services, and I do not have a RAID environment. I suspect most folks do not have a RAID system.

Originally Posted by OreoCookie 

I haven't omitted that at all, for instance, I wrote that »Clones of your hard drive can help you minimize downtime in case of a hard disk failure (you can work off the clone, for instance).« Also I put emphasis on this when I quoted the FAQ from the developer of SuperDuper (last emphasis to be precise). If you intend to use the cloned drive to cut down your down time in case your primary drives no longer works, then that's exactly what clones are for.

You did omit mentioning the serious defect in Time Machine (now looks like there are some others, based on some of your statements).

Originally Posted by OreoCookie 

Time Machine is not flawless (I'm particularly worried about HFS+ corruption and bit rot), but it's not the only backup tool I use -- and it certainly is a lot, lot, lot better than anything that existed before. And it's free. It's not a defect of Time Machine that you can't boot off your Time Machine drive, because that's not what Time Machine was designed for -- Time Machine was designed to be an easy to set up incremental backup, and not a cloning software.

Super Duper was designed to do an accurate backup, and allow one to easily and quickly do a restore. Also, one can use Super Duper in "sponsored mode". What one gives up with that, of course, is automated backups. One must do their backups manually. But, that is fine with me, as 1) I control the backup processing, and 2) I have been doing manual backups for so long, it is second nature to me..

Originally Posted by OreoCookie 

Launch Activity Monitor and look at all the processes which are running even when you have no apps running: it's a lot (make sure to show all processes). With modern backup tools such precautions are no longer necessary because they have been designed for »modern« multitasking operating systems (I write modern in quotation marks, because it was modern 20 years ago). There are backup tools for databases, for instance, that can be used while the database is running.

I am aware of those, but I believe I am "isolating" my environment enough for my backup purposes.

Originally Posted by OreoCookie 

Look, if you just came here to defend your way of doing backups with SuperDuper, then let's stop here. But it should be a clear sign to you that not just all of us here who participated in this thread, but also Anandtech, SuperDuper's software developer and the guy whose post you linked to also use and recommend Time Machine -- possibly in addition to cloning software, but not the other way around.

I am not necessarily defending Super Duper. I really thought my initial post was fair and balanced. I have yet to say that one method is superior to another, and that my use of Super Duper is fine for my needs. Others, like yourself, have, in some instances, put both me down and Super Duper. For example, in your post on July 16th, you state:

"@akent35
I'm not exactly sure why you're posting here, but let me at least answer the two questions"

Who are you to say that? I have never said anything remotely like that to anyone.

And, in your post yesterday (7/17), you state:

"But clones by themselves are a very bad backup solution, because you don't have access to older versions of your files"

I have yet to use the word "bad" in reference to Time Machine. But, it does have a serious defect, in not being able to boot from it. Yes, that can be overcome, so I am not saying it is very bad. I am also not saying it is bad. There is nothing very bad about Super Duper.

[mattyb]

Originally Posted by prl99 

These are accessible from the internet but aren't necessarily hosted on cloud servers.

What's a cloud server?