Skip to main content

"Cannot find version 9" - stopping backup of version 12!!!

Thread needs solution

I run Acronis True Image 2016 to backup my PC to a NAS drive (QNAP TS-412).  I maintain a number of full backup generations as well as 3-4 incrementals per full backup.

I have a problem which crops up over 3 months or so whereby ATI attempts to perform a backup and then complains about not being able to find a version which is a number of generations back.  The files are still there on the NAS and are accessible.  In my current situation, have a full backups of versions 5 through 12 on my NAS but it is complaining that it cannot find Version 9.

Now I can understand that it is possible that an older version such as 9 might become corrupt on the NAS drive.  However (and this is the bit that really irritates me...), why should ATI even care about looking at version 9 when it is trying to perform an incremental on version 12 or a full on version 13?  It isn't even that it is the next version to be deleted (that would be version 5).  Worse still, it won't even just issue a warning ("Hey, I notice version 9 is corrupt, but don't worry, 10, 11, 12 all look okay!"), instead it refuses to conitnue to do ANY backups until version 9 has magically been fixed (if indeed there is a problem with it).

Also, the dialogue box says "Please click Browse and specify its location, or ignore this message".  There is no "Ignore" button, just "Retry", "Cancel" or "Browse...".  Browsing to the file just brings the message up again after 5 seconds.  Cancel cancels the backup attempt.  Retry is the same as browse.  How does one ignore the message but allow the backup to continue?

I am getting fed up with having to delete down and completely recreate my backup strategy when this happens.  Also, this problem existed in ATI 2014 too.  I was hoping it would have been sorted by now.

If anyone can provlde a logical answer to the question in red above, I would be most appreciative.  Better still, if anyone can suggest how to get around this problem (e.g. - tell ATI to delete/forget version 9 from it's repository so I can continue to do normal backups) then you will be my new best friend!

NOTE: I've checked disk space and NAS Share permissions.  Both are okay.
Thanks,
Chris

Attachment Size
cannotfindversion9.png 4.69 KB
version9.png 57.29 KB
0 Users found this helpful

Perhaps your picture did not include them but several files (both full and incremental) appear to be missing.

The program may also be referring Inc B11_S9 rather than Full B9

If there is a fix possible, it would be for you to run a manual validation and respond "ignore" to each prompt and may require many "ignore" responses.

In order for a backup to be Full/incremental backup series to be valid, all pieces (slickes)  must be present. None can be missing.

The BXX S1 is the first and every subsequent S# must be sequential and accounted for.

Each incremental backup contains a siingle data set and not data is NOT repeated in any other incremental backup.  If an incremental is missing, then a piece of the backup is missing and the backup will not create any further backups.

At this pont, your corrective action is to  create a new backup task which has a new task name and I recommend that you also point the new task to a new storage folder or sub-folder so the  old and new backups are not intermixed.

Custom Backup Schemes
https://forum.acronis.com/forum/100416

GH25. Understanding differences between Incremental and Differential backups for data recovery.
GH12. Create Custom Incremental Backup Scheme.
...Keep Full plus 6 Inc per chain. Store/Keep 4 chains. Use whatever number best fits your needs.

Thanks for the response GH-Storage.

 > Perhaps your picture did not include them but several files (both full and incremental) appear to be missing.

Sorry, I should have included a picture of the entire respository - NOTE that I've listed it in DateModified order (most recent at the top) and it is showing the reepository as it is "as of today" (attached).  

Somehow I managed to get the backups kick-started again yesterday morning and it is done a full backup yesterday and a full backup today (not quite what my backup configuration is but there it is).  It appears to have decided to start backing up from B9 again even though I as prevoiusly at B12.  The B9 corresponds to the error message message I was getting ("could not find version 9").  Because I already have B9/10/11/12 backups in the repo, it has started attaching "-2" to the filenames.

The program may also be referring Inc B11_S9 rather than Full B9

Wow! I've never considered that but that is worth considering.   I am currently assuming that "version 9" is the B number (generation of Full backup) and not the S number because since the backups have decided to start working again, it has done two full backups as full_B9_S1_V1-2 and full_B10-S1-V1-2 rather than full_B12... or inc_B12_S9...

> If there is a fix possible, it would be for you to run a manual validation and respond "ignore" to each prompt and may require many "ignore" responses.

This isn't something I've tried.  My config is to a validation "Once a month".  By the way, my Backup Scheme is:

  • Incremental
  • Create a full version after every 9 incremental versions
  • Cleanup = Keep size of backup no more than 2TB (enough for lots of full + incs and NAS has 2.8TB of space)
  • Schedule: Mon, Wed, Fri at 09:00

I will try this option next time.

In order for a backup to be Full/incremental backup series to be valid, all pieces (slickes)  must be present. None can be missing.

By "slickes", you mean "all S numbers under a given B"?  That is how I understand it on the basis that S=Incremental in my current backup scheme.  I still don't believe prevoius full backups (previous generation of B numbers) should be required by as per your first comment, it is ambiguous as to what "version 9" refers to "B" or "S".  S would make more sense, if it hadn't started backing up again at B=9.

> The BXX S1 is the first and every subsequent S# must be sequential and accounted for.

Agreed.

> Each incremental backup contains a siingle data set and not data is NOT repeated in any other incremental backup.  If an incremental is missing, then a piece of the backup is missing and the backup will not create any further backups.

Agreed.

> At this pont, your corrective action is to  create a new backup task which has a new task name and I recommend that you also point the new task to a new storage folder or sub-folder so the  old and new backups are not intermixed.

This is what I've had to do in the past and it is a pain in the neck as it means going through all the config options again and manually deleting down the old repository over the subsequent 2-3 months - and I've been gaving to do this on a six monthly basis when this error occurs.

If there is anything identified as corrupt about a previous full or incremental backup, then I would expect any sane backup software product to increment the B number (Acronis convention) and perform a fresh full backup without any fuss.

> Custom Backup Schemes.. https://forum.acronis.com/forum/100416

> GH25. Understanding differences between Incremental and Differential backups for data recovery.

> GH12. Create Custom Incremental Backup Scheme. 
> ...Keep Full plus 6 Inc per chain. Store/Keep 4 chains. Use whatever number best fits your needs.

I'm familiar with backup schemes concepts as it is part of my day job.

Thanks,
Chris

 

 

 

 

 

Attachment Size
323712-125125.png 102.24 KB

Chris wrote:
If there is anything identified as corrupt about a previous full or incremental backup, then I would expect any sane backup software product to increment the B number (Acronis convention) and perform a fresh full backup without any fuss.

Unfortunately, the action that would be taken by 2016 or before would be to stop with an error message.

Acronis namiing excerpt: wrote:
If you are creating a new backup, and there is already a file with the same name, the program does not delete the old file, but adds to the new file the "-number" suffix, for example, my_documents_inc_b2_s2_v1-2.tib.

There are 5 instances of this happening in your backups and we don't why--except perhaps the consolidation caused by the scheme setting "keep size of backup no more than 2TB".  There have been past reports of the program losing track of the consolidated backups with the not able to find etc... Your log file on those dates may provide some information. I assume you have email notification enabled so you get immediate notifiction of success/failure of each backup at time of running.

Should you need or want to create a new task in the future, I would recommend that  instead of using scheme option "keep size of backup no more than 2TB", that instead, use the scheme option "store no more than 4 version chains", plus enable the option "do not delete" original.

As you seem to average 300GB for each chain of full plus 9 inc; your space requriedments would be

4 keep x 300GB=     `1.2 TB

1 Full keep original =  275GB

plus 275GB  temp need while each full is being created before deletion of oldest chain.

Those settings would need about 1.8GB.  If you wanted to keep more chains, then increase the "store no more than 'n" setting. Each additional chain would need about 300GB.

Backup scheme setting
     Custom/Incremental

     Full plus 9 incremental

     Store no more than 4 recent version chains

     Enable Do not delete original.

---------------------------------------

My reasoning for avoiding the use of a disk limit is the merging or consolidation.  Use a set number of chains to be stored provides info so  you know exactly when that deletion of oldest chaiins will occur and you know exactly what (1 chain) will be deleted each time.  Keeping the original (do not delete original) is outside any automatic deletion. This setting would cause Full B1 to be kept until you manually delete it. Full B2 would begin first chain subject to automatic deletion.

I will I could offer a good reason for your naming issues. The usuaal cause is the target is not available or a temporary loss of target connedtion.

I would consider setting the target to a "no sleep" setting, or a a method of making sure the target is seen before the backup begins.

Good luck.

Just to say have same problem. An my NAS is availble. Integral Full every 5, clean after 14.

I often got the error, while working, so I check imedialtey local LAN, all good.

I never had such with Norton, Symantec Ghost.

For the price I expect smooth operation and self healing. If not fund, start new verison.

I am constantly getting this error message during validation of the incremental backup. The drive being  used for backup is fine according to chkdsk and all versions are present. I've used 2 different folders on the same drive and get the error, but with different versions eg version 1 instead of version 5,6 or 7 in the other. Browse doens't work. The only option each time is to start a new backup, but, if you can't rely on the backup programme, what is the point of having it?