Replacing an incremental backup

I'm not sure I understand what you are after. You can certainly make an incremental back, and along with the Full and nay previous incs in the set, it will be capable of restoring your drive to the condition it is in today. In fact, you could, for example, make an inc everyday for, say, a week, and be able to restore your drive(s) to the condition it(they) was(were) in on any one of those days.

If you only want to have one "change" file associated with a Full, then you want to make diffs, not incs.

Slap Happy wrote:

Situation:

Questions:
1. Ideally I would like to create a new incremental backup off of the original full backup. Is this problematic? Can I make a second incremental backup off the original backup so that effectively I will now have two incremental backups that are both incremental off the same original backup?

Yes

My intention would be to delete the first incremental backup i created once the second one is created.

You cannot delete a intermediate incremental, all previous incrementals must be present to enable a restore to be carried out

2. Assuming I can make this second incremental backup off the original full backup, by having first restored my partition (in step 3 above) will the new incremental backup now be just as large as the original full backup? I have seen it said in the forums "An incremental or differential backup created after the full one was restored will be considerably larger than usual. This is because restoring a backup archive changes file locations on the disk, and backups reflect these changes."

Maybe I only use full backups, try it you can always delete it afterwards

Although this is not exactly my case (i did not restore from a full backup, but from an incremental one) I am wondering if its would still apply?

The restore always restores the full backup and the selected incrementals

3. Any other advise for this situation is welcome!

You can make a new full backup, you don't say where you are backing up to, if space is at a premium then you could delete the previous backup and incremental

Thanks very much for the feedback, but I don't think i explained myself clear enough. I will try again to ask my main question.

I have a full backup and an incremental backup (call it "INC#1"). What I would like to do is ignore INC#1 and make a new incremental backup ("INC#2") based off the same original full backup. I want the new incremental backup "INC#2" to be completely independent of the first incremental backup "INC#1", so that after i create INC#2 (and verify it) I can delete INC#1.

Is there a problem with doing this (e.g will it automatically try to make the new incremental backup up ("INC#2") dependent on "INC#1? Would I need to delete INC#1 before I create INC#2 to have INC#2 only dependent on the full backup?

You can do that in versions of ATI prior to ati2009, as the full has no idea how many incs succeed it. In later versions it will work but it might make the ati database go a bit wonk expecting the old inc to still be there. so I'd remove the inc, then open ATI and try to find/explore the inc form within ati -- it will generate an error but update the database. Then make your new inc. In any event, you'd want to remove the old inc before trying to make the new one.

the new inc will contian all the sectors that were written to since the first full was made.

Thanks Scott for your quick response!
You understood my question perfectly. I'm glad i asked, I had a feeling it could cause some problems.

you said:
the new inc will contian all the sectors that were written to since the first full was made.

so am i correct in assuming that because i have just restored my disk to the saved state of INC#1, made a few small changes, and if I now delete INC#1 and make a new INC backup (to include the few small changes i made), the new INC backup will be almost as large in size as the full backup (because of the disk restore that i have done)?

If you did a normal restore and TI was able to put things back where they were, the Incremental should still be small. You would need to run it to find out. If it's large and you really want a backup in this state, I would recommend creating a Full instead.

Slap Happy wrote:

Thanks very much for the feedback, but I don't think i explained myself clear enough. I will try again to ask my main question.

I have a full backup and an incremental backup (call it "INC#1"). What I would like to do is ignore INC#1 and make a new incremental backup ("INC#2") based off the same original full backup. I want the new incremental backup "INC#2" to be completely independent of the first incremental backup "INC#1", so that after i create INC#2 (and verify it) I can delete INC#1.

Is there a problem with doing this (e.g will it automatically try to make the new incremental backup up ("INC#2") dependent on "INC#1? Would I need to delete INC#1 before I create INC#2 to have INC#2 only dependent on the full backup?

In all versions of TI that I have used (since version 8) incremental backups have been reliant on the previous increments. Each increment being a backup of the changes since the last one, thus if you have a chain of increments let’s say 30, and increment number 3 was removed or became corrupted for some reason, you will only be able to restore the full and increments 1 and 2. This is the reason that I and some other members of this forum only perform full backups.

In a recent version I can't remember which Acronis introduced "Differentials" these are not reliant on previous differentials only on the original full, however in the first incarnation of differentials the last differential could not be validated and therefore restored unless all previous differentials were present in the same folder.

I enclose a copy of an extract from the user manual for TI 2010, as I don't have 2011, I don't know if the situation has changed. May I suggest that you download and read the user manual for the version you are using.

Quote from UM 2010
3.2. Full, incremental and differential backups
Acronis True Image Home can create full, incremental and differential backups.
A full backup contains all data at the moment of backup creation. It forms a base for further incremental or differential backup or is used as a standalone archive. A full backup has the shortest recovery time compared to incremental or differential ones.
An incremental backup file only contains data changed since the last backup of any type (full, incremental, or differential). Therefore, it is smaller and takes less time to create, but as it doesn't contain all the data; all the previous backups and the initial full backup are required for recovery.
Unlike an incremental backup, when every backup procedure creates the next file in a "chain", a differential backup creates an independent file, containing all changes since the last full backup.

Generally, a differential backup will be recovered faster than an incremental one, as it does not have to process through a long chain of previous backups.
End of quote

Thanks MudCrab and como for you feedback

However something that confuses me about the second part of this statement.

como wrote:

In a recent version I can't remember which Acronis introduced "Differentials" these are not reliant on previous differentials only on the original full, however in the first incarnation of differentials the last differential could not be validated and therefore restored unless all previous differentials were present in the same folder.
End of quote

You are saying that if you create more then one differential off the same full backup you need to keep all the differential backups to restore from any of them?

No, the first version of TI that had differentials as an option had an undocumented feature (BUG) that you had to retain all the differentials in the same folder to enable the validation of the last differential and full backup. This defeated the object of having differentials, as each file was larger than an increment, some users found that their disks were quickly filling up!

I believe this was fixed in the next version

Not the first version but nearly all versions of ATI required that all previous diffs in a set be present to be able to validate a diff but not to restore. It was just a quirk in the Validation coding. Might have been fixed in the later versions, someone else willahve to confirm as I don't use diffs.

I only do full images of my system drive, all my data which is on a second internal is backed up by a free program that backs up files in their native format, so no first hand knowledge of differentials, but do remember many posts re this on the old forum