Differential backup scheme doesn't work properly

Please post a screen capture of backup scheme.

Attached. Is this what you need?

It looks like the backup task is failing to see the existing backup. Make sure you don't manually rename or move any file that ATI is creating. Try to set up a new task.

Also, do you have the computer to run missed backups at startup? That might be why you see it try to do all these backups.

I believe you consolidation picture of keeping 4 versions is saying that it will do 4 full and then start the diff cycle of consolidtion. Did you confirm the name on the 2nd backup as not being dif?

My suggestion would be to stop using the current task and change to a custom/differential backup scheme which uses automatic cleanup.
Note the example below. First, understand that TrueImage will not delete the old backup until after it has created its replacment first. Therefore, you have max 1.8G to use but the extra temp full means you have 1.6 tied up so you only hve 200 GB for use for diff type files.
So, you really do not have much space for diff files. When the next full is due, one option might be to use the program and delete the existing backup. If you were to choose this method, you could plan on creating another 800GB of diff before forcibly deleting the old full plus all the diff.

TrueImage backs up used disk sectors so be sure and no due any auto defragging prior to any backup as doing so will cause major increase in size of backup.

I would encourage you to add additional backup storage as it is never good practice to delete your only backups.

Figure 11-Dif: -Example of custom/differential backup method settings

If using differential type backups (which is full + X Diff), the 11-Diff illustration is my recommended method. Change the 6 or 4 to fit your available storage.
These automatic cleanup settings will provide for automatic deletion of the oldest backups after the "Store no more than X number of chains" quota has been reached.
In this example, deletion of oldest backup chain will occur immediately following creation of backup #29.

Allow space for 1 more full backup in addition to "Store no more than X number of chains" as the program will NOT delete the oldest full until its replacement has been successfully created.

In this example, one chain or one recent version chain =1 full plus 6 diff or 7 files per chain.
If keeping 4 recent version chains (4 chains of 7 each) retention would be 28 files. Deletion begins after backup 29.

When restoring, a single diff file is selected and that specific diff plus its full backup base will be restored (2 files restored).

Grover,
Thanks for your detailed reply. I was going to check my settings and implement your suggestions, only to find my True Image is busy doing god-knows what this evening.

First, to answer your questions, on the prior run it logged an error attempting to find the full backup, then proceeded to overwrite it in the exact location where it was looking for it, with a new full backup. It certainly did not wait until a new full backup was completed before overwriting the old one.

I deleted the old file backup settings and had it doing a partition backup. It did the first full backup just fine. Last night at 1:00 AM it ran the first differential backup and as expected, only created a 721MB file. (Hurray!) However it then decided it's time to consolidate, even though there were not 4 differentials yet. The screen capture below shows what's going on in the target directory. There's the initial backup of 812gigs, a differential of 721 MB and 2 unknown files I can only assume are a work in progress of 139 + 591 gigs. Clearly there will not be enough room on the drive for whatever this program is trying to do with itself for nearly 24 hours. From what I understand based on my settings it should have done 4 differentials before consolidating.

The other mystery is as you can see from the screen capture the program is not doing anything right now with the My partitions backup. It is not scheduled to start for another 2 hours. However the backup drive is flashing and the program is doing I/O to drive F: for no apparent reason. It is scheduled to do a validation at the end of the month but should be idle at this hour.

I've been doing this for a couple of decades now but have never encountered a program this frustrating.

Update: as of this morning the destination location for the N backup contains files from the "My partitions" backup. So it seems the program was actually consolidating the My partitions backup last night, not the Data (N) backup as it was showing on the screen, to the wrong drive. So is it possible that the program is confusing the tasks for the 2 partition backups?

I attached screen captures showing the destination folders for both backup jobs.

Well I never intended to bore everyone with a running account of my exciting backup saga. But after running a few cycles I've concluded the program is buggy. To recap, here's what happened.

On day 0 I created job #1, to backup drive C and D to drive F. That job ran fine for a number of weeks without issues till I created job #2
On day N Job #2 was created to backup drive N to drive M. I launched an immediate backup. It ran fine and created a full backup.
On day N+1 job #2 differential ran. It could not find the full backup created the prior day, but proceeded to overwrite it with a new full backup with the same name on drive M. The log said it was looking for the backup in the correct place, with the correct name, but clearly it was not.
On day N+2 job #1 differential ran. It wrote a differential file to drive M in error, then consolidated it on drive M in the folder reserved for job #2. The differential should have gone to drive F.
On day N+3 job #2 differential ran correctly and created a small differential on drive M.
On day N+4 job #1 differential ran, could NOT find its full backup although it reported, once again that it was looking in the right place and had the right name. It created a new full backup where it belongs on drive F.

Conclusion: The program corrupted the settings for destination folders for jobs 1 and 2 when job #2 was created.

What will happen tonight is anyone's guess but I'm hoping the program has figured out what to do with itself from now on.

Please find and fix this bug.

I can only imagine that the internal database of ATI is corrupt. Let's try this.
Delete your current tasks. Then look for archives.xml in the C:\programdata\ACronis\Trueimagehome\database. rename the file archives.old.
Launch ATI and recreate your tasks.
Does the problem persists?

Jeff,
Is it posible for you to post a screen capture of each backup scheme used on the bad tasks?
If no, can you tell us more about the actual scheme used.
Was it custom?
Was it automatic cleanup & how many "recent x recent version chiains"?
Was consolidation involved?
Were the tasks original or had they been edited?

We want to try and find out what did not work,

Thanks I'll keep that in mind in case it happens again. However for now the problem seems to have repaired itself. Last night's differential backup (Job #1) ran fine.

This "delete the jobs and recreate them" thing seems to be a recurring theme on this board. Usually that's a good indicator of an unstable program that's not able to manage its own data structures.

Jeff,

If you have the time, can you share your setup as mentioned in my post #9. It would help us to have a better understandidng of what is happening. Thanks.

Grover,
I attached a bunch of files. Hope this helps you find the problem. I can send in the logs as well but I don't want to post it in a public forum.

Main screen (Job #2 failed because the destination folder was full of stuff put there by job #1)
Job #1 setup screen 1 & 2 and destination folder.
Job #1 log file shows it deleting a file on drive M (Job #2's folder)
Job #2 setup screens 1, 2 and 3 and destination folder
Job #2 log file shows it deleting files on drive F (Job #1's folder)

Glad to help you get to the bottom of this. Let me now if you need anything else.

Yes, I am reviewing our attachement. What is the size of the storage partitions where your backups are stored?

It would be helpful if you were to post a screen captue of your Windows Disk Management graphical view.

Both destination drives are 2 tb. I'll post those details later.

The task has not repaired itself. The following notification was received

DATA (N) Task is waiting for user interaction.
Description: Stage Description
Information: Failed to open backup M:\Backup_N\DATA (N)\ltbkp2014-01-10My partitions84_diff_b14_s2_v1.tib. Make sure the backup location is accessible and contains all versions of the backup. This backup may also be corrupted. 
Details: Click Retry to try to read from the same location. Otherwise, click Cancel to cancel the operation.

This is job #2 looking for job #1's file in job #2's location.

Both destination drives are 2 tb. I'll post those details later.

The task has not repaired itself. The following notification was received

DATA (N) Task is waiting for user interaction.
Description: Stage Description
Information: Failed to open backup M:\Backup_N\DATA (N)\ltbkp2014-01-10My partitions84_diff_b14_s2_v1.tib. Make sure the backup location is accessible and contains all versions of the backup. This backup may also be corrupted. 
Details: Click Retry to try to read from the same location. Otherwise, click Cancel to cancel the operation.

This is job #2 looking for job #1's file in job #2's location.

Thanks for the info about the 2tb disks. I do plan on responding but time is short right now.

It would be helpful if you were to post a screen captue of your Windows Disk Management graphical view.

Disk manager screen attached.

Job 2 was attempting to consolidate when it sent me the message in post #15 above. It's set to do only differential versions after the initial full version and consolidate when size of backup exceeds 100GB. However the backup was only 12 GB as seen in screen capture #2 below. I attached the log for the run as well.

Thank you. I will try to respond tonight.

Hi Jeff,
I have reviewed your attachments and offer these comments and suggestions based on my experiences with the program and comments posted by user on the forum. Understand, I am not a tech specialist but a user helping another user.

1. First, I have rules for myself which I follow and suggest that others follow as well. These are my personal rules and some contradict what the program allows the user to do.

A. I try to create the task in its final form the first time. Except for a change of schedule, I will NOT edit or make changes to an existing task. (Note: my rule contradicts what Acronis permits.) If changes needed, I stop using the task and create a new task plus point the task to a new destination folder or sub-folder. No mixing of different task backups into the same folder (Note: my rule contradicts what Acronis permits.)

B. I will not use any of the standard default backup schemes. The defaults do not provide for the proper user control of backup retention. All my schemes are custom with automatic cleanup. Automatic cleanup uses a straight forward simple delete the oldest backups method and there is no consolidation.

C. For my backup schemes, all use the automatic cleanup plus the option "store no more than X recent version schemes. If using full plus inc or full plus diff, about 15 is the max number I allow myself before a full must be created. I feel too many backups dependent upon a single full is not a healthy situation for the recovery of data. Having too many inc's is a mistake frequently posted on the forum where recovery is not possible due to one of the backup files being corrupt or missing. Some users have posted having as many as 80 or even 800 inc's and then wonder why their backup has problems and a restore was not possible.

D. For my backup schemes, I will not use any of the options based on
"Delete chains older than X days" or
"Keep size of the backup no more than X GB
"Size of backups exceed X amount".
Your tasks use these option and I discourge their use.

Perhaps I am paranoid but I simply will not allow the program to merge my data and all 3 of the just listed option is a form of consolidation. Consolidation is certainly offered as options for the user to choose but any consolidation requires the programmer to be perfect and your problems of mixed consolidation points out is another indication as to why it should be avoided.

2. Specific to your postings.
a. Job 1: The backup type is a partition backup which does include all partitions. I suggest you change this to be a more inclusive "disk image" which involves for you one more step which is the click the "Switch to disk mode" option. You are already including everything but this type image can make for easier restores without any loss of restore features. Restoring a single image is still possible from the "disk mode" or "disk image" backup.

b. Job 1: Storage disk is F which is 1.8TB available.
Scheme set to full after 5 diff
Keep size to 1100 GB.

Full disk backup is listed as 622 GB. When TI needs to create a new full, you need another 622GB (temporarily) before any deletion occurs so you have about 600 GB available for storages of any other inc or diff type backups. You really do not have enough room for 2 full backups plus the additional temp space for next full before the old full deleted.

One option for a new scheme would be to have a custom/diff scheme using automatic cleanup (no consolidation)
--full after 15 diff
--Store no more than 1 recent version chains.
Note: you do not have enough room to store more than the 1 recent chain but you do have enough storage space to increase the number of diff's kept but keeping too many diff is not datasafe.
This setup would basically provide for 1 full monthly. Or, if you wanted twice monthly full, you could reduce the 16 to 8. Unfortunately, you only option to make the next backup a full backup is to control the number of Diff backups.

3. Job 2:
b. Storage disk is M which is 1.8TB storage available.
Backup scheme is custom/diff with the only limit is keep size to 100Gb. which involves consolidation.
Full backup size is 813. GB. Temp space needed when a next full is created is 813GB so you really only have 280 GB or so for diff backups in between the full backup creations.

A new task could be created with the same settings as Job 1 so you would set
1 full plus 16 diff.
Store no more than 1 Recent version chain.
This scheme would consume *813GB +813GB (temp) plus 200GB for 16 diffs or about 1.75 consumed in retaining 1 full plus 16 diffs--assuming that the amount of diff size changes is no more than 200GB for 16 diff backups. Your dif size really jumped from Jan 13 to 15th. Hopefully, the remaining diff would not be any larger in their jumps for a single backup.

This setup would basically provide for 1 full monthly. Or, if you wanted twice monthly full, you could reduce the 16 to 8. Available space is could get very close to a "full disk" error message when the next full is attempted. On the day you create the next full, you need free space of at least the size of the approaching full backup as the next full must created before the old full is deleted by the programs.

Perhaps you have some static data on Drive N which could be moved to another disk so the size of the backup would be reduced. Drive M is marginal for keeping 1 full plus 16 diff on your current Drive M due to continued need for the temp storage space before the old full is deleted.

As the month end draws near and all backups being completed, there will he no room on Drive M for a Win7 virtual box or the disk full message will appear before all backups stored on M.
-------------------------------------------

Bottom Line changes recommended
1. Cease to use the old tasks.
2. On Job 1, Change backup to a Disk mode backup.
3. On both Job 1 and Job 2,
Create new backup task with custom/diff scheme with settings of
Automatic cleanup enabled.
Full after 16 diff; or full after 8 diff
Store no more than 1 Recent version chains.
No consolidation.

----------------------------------------------
As to the cause of the mixup during consolidation, that had to be a problem with the program. If you started over with fresh task but with the exact same option settings, I believe you would have same problem again--which is why I suggesting changes but changes which I have used and work.

When Acronis released version 2014, they indicated considerable time has been spent on making the backup retention settings more dependable. Perhaps the code for consolidation was improved but I don't know. What I do believe is that the setting I have recommended do work in 2014 and "should" work in 2013 if you follow my suggested rules.

Grover,
Thanks once again for the detailed response. I implemented your suggestions for backup #2 and will likely fix backup #1 when time permits or it starts to cause problems. I also freed up space on the drive so there's room for 2 full backups.

I believe your post has validated a lot of the concerns I had. This program is buggy. It's a wonder that these issues were not found and fixed in beta testing since it's obvious that documented features simply don't work. I have an uneasy feeling entrusting my precious data to such a flaky program.