Cleanup
Moved from ATI 2019 to ATI 2020 little over a week ago on my WIN 10 PC. Simply removed old backups and created my two new backups: One is the Main Windows 10 drive, SSD card on MB > SSD internal drive; Two is a Revo SSD card > same SSD internal drive....see below pic.
When using ATI 2019, I used the Options shown below and the old full back plus six differentials would be removed upon completion of a new full backup.
Just watched to see what would happen with the new ATI 2020 backups with this same cleanup option: Did a full with One (SSD M21) backup listed as 0007.tibx and left several former backups in place after several minutes of cleaning up all but the former full as an very small file...but left it. The Two (Data Revo) did a 0007.tibx full backup but left all prior differnential plus the full backup from a week earlier... .tibx.
Q. What backup options should I be using to allow just one full backup to remain for each of the two backup schemes and have the prior full and differentials removed?
Thanks!
Bob
The files that finish in the -000x.TIBX pattern are files related to the main TIBX file. After each differential run, the chain of TIBX files is updated, the last -000x.TIBX is modified, and a new -000x+1.TIBX is created, even if it is a new full that just ran.
As you saw, when the second full gets completed, the chain shows an initial TIBX file that is very small, and -000x.TIBX that are larger. As the chain progresses, the small TIBX file remains.
The key here is to understand that the -000x.TIBX files are not the same as the different volumes and slices vYsY naming convention in previous versions.
Therefore, do not interfere at all with the creation of the backup files.
Whenever you want to restore using the recovery media always select the main TIBX file as the archive to recover. If you select a dependent file, you will geta deceptive "backup file is corrupted".
Ok...got it....but wondering why the second backup job in the same folder did not delete the diffferential files? The first did ok and took several munutes and did leave that base .tibx file.
Maybe I did not have it setup in options correctly? Will see next Sunday.
I copy the full backups to a external hardrive manually after these full backups occur.....but it appears these are of no value by themselves? Do I need to copy the base .tibx each time as well in order to get something off these full backups?
This is a good question. With this new TIBX format, I don't know what is the minimum set of files you should manually copy to keep a chain valid. Since there is no easy way to look at the backup files and determine which one of the dependent files if a full or an incremental, aside from looking a the size.
But even then, I am not sure that selecting only the base .tibx file and a valid -000x.tibx full but leaving, let's say, the subsequent differentials would result in a valid chain at restore.
Maybe you can try and share you results with us.
I would err towards copying/synchronizing all the files.
If you are trying to protect your backups, I think it is better to do another backup of your source data using the ATI reserve copy, another ATI task or, better, some another technology (maybe a file backup Windows 10 file history or some other synching software like SyncBackSE) to another destination, ideally off site.
I copy full backups and can access any file I may want by openning that backup and copying that file....may be several months old.
Never have had any need other than that.
I use Casper to keep my main win10 drive cloned daily to another ssd...m22.
Maybe I will go back to ATI 2019 or use Win backup?
Thanks!!
Bob
BOBBY wrote:I copy full backups and can access any file I may want by openning that backup and copying that file....may be several months old.
Sure, that is the way it was working with regular .tib files. With the new .tibx files, things are working differently.
I did some testing on my computer and I couldn't find a way to copy the tibx files and alter them to keep only the fulls as you are trying to do.
I started by creating various altered copies of the backup folder containing the base reduced TIBX, one differential starting with a -0014.tibx big enough to the full, a few differentials depending on -0014 and yet another full (-0018.tibx)
I made various copies without or without the main TIBX, deleting only differential attached to the -0014.tibx, or deleting the whole -0014tibx chain and keeping only the last full. I tried renaming the files...
Anyway, I was not able to create a subset of the files and add them as a new backup. Everytime, ATI was telling me
'Failed to add the backup to the backup list. The backup may be locked or corrupted. Also, make sure the folder contains the last volume of the backup, and does not contain a renamed copy of the same backup.'
Clearly, the base TIBX file contains information about which subsequent backups have been executed. Clearly, you cannot add a backup pointing at a with -000x.tibx, it has to be the base TIBX file.
Eventually, the only copy that worked was a clone of the folder with all the original files.
I'd stay with 2019 then, if you want to continue manipulating the tib file chains manually.
Appreciate, the work on checking if keeping the full backups would work as I have done in the past.
These full backups I kept via ATI 2019 were kept on a 2 tb hdd and offsite and updated weekly. Good for full restore if necessary as well as retriving files from them if needed.
I guess I can't do that with ATI 2020, so back to ver '19.
What with this thread as well as the issues related to validations being run on all chains in a .tibx backup (as opposed to just the latest), I think we really need to get some clear guidance and understanding from Acronis as to the way in which .tibx chains are handled. It seems as though there are a lot more interdependancies of files than was the case with the .tib format. As it is now, there are too many ways to get into new trouble.
BrunoC wrote:What with this thread as well as the issues related to validations being run on all chains in a .tibx backup (as opposed to just the latest), I think we really need to get some clear guidance and understanding from Acronis as to the way in which .tibx chains are handled. It seems as though there are a lot more interdependancies of files than was the case with the .tib format. As it is now, there are too many ways to get into new trouble.
+1 There are many aspects of the new backup architecture which are not documented, making it (unnecessarily) difficult for users to get the outcome they want.
Ian
Well it appears that ATI 2020 is working and I am able to resolve individual files via both Acronis and File Explorer!
I copy the full backup files to an external 4TB usb drive and the name is changed that includes the date of the backup.
In the pic attached, you can see from Acronis the two backups and the two copies on the external drive. From Recovery, I can see the individual files from either drive (local or external). The local are in the Blue rectangle and the copies in the Red rectangle. The File Explorer is shown just below.
So, before the Full backup on each Sunday, I will open Acronis and do a Cleanup on each of the two backups which will remove the full and differential backups from the prior week...the full will remain on the external drive for use as needed (automatic cleanup is Turned Off). Then the new full backup will result and be copied to the external drive (with date in the name) which can be accessed. I can access these backup copies either in Acronis or via the File Explorer.
Yes, I use first ATI version cleanup to reduce your differential backup chains to the full(s) that you want, you will end up with a valid chain that can be copied away.
I can see 3 major drawbacks of this:
- First, each week you are losing the differential backups of the week unless you first do a copy of the entire "original" backup chain to the USB (a bit long, but manageable if your files are not too big), add this new copy to ATI and then clean it up using ATI,
- Second, you end up with a "long" list of other backups stored on the USB disk
- Third, this is very manual and if you have track these full backups manually.
I would rather set up 2 different tasks in ATI: one with differential backups, one with full backups only to the USB.. With the TIBX file format, full backups can be faster than before.
At any rate, it looks like you have found a way to achieve what you wanted.
Good idea.....I will just do a Full followed by several weeks of Diff backups; save the Full to the External USB (automated with Macro Express that adds the date).
Then when the Backup Drive gets full I will do a Cleanup and start again.
Thanks!
BOBBY wrote:[...] save the Full to the External USB (automated with Macro Express that adds the date).
[...]
Since you are moving backups around, always check that the result of your copy is a good chain.
A quick way to verify you have a good chain is to try to open its TIBX file. If ATI complains, you wouldn't be able to restore and your copy would be useless. A more thorough way is of course to validate the chain (but validation is a longer process with 2020), and ultimately, the best way is to attempt a real restore. The Windows explorer check should be good enough to ensure your copying of files has not messed up the chain, however reduced it might be.
Good luck
BOBBY wrote:I copy full backups and can access any file I may want by openning that backup and copying that file....may be several months old.
This comment is a bit off the Cleanup topic, but I think it is important enough to make here.
Several people have remarked that saving old full backups is an important part of their backup management. I think it is important that Acronis knows (and understands) that they have removed an important feature. There needs to be a way to save the smallest instance of a backup chain - whether that be the full or a consolidated version of full plus incrs or diffs.
Maybe if enough of us submit feedback Acronis will provide a solution. And if this capability is already available in the new format, Acronis needs to better document it.
+ to this Patrick