Continuing Backup Failures

I do not recall having a similar problem when backing up to my Synology NAS. 

First off, am I correct in assuming for the time stamps that the tasks run overnight; and if so, are the computers hibernating or doing a cold start? What model Synology do you have and what is the total storage and how much free storage. What cleanup rules do you use for the backup tasks?

I am uncertain about what is included in the backup tasks; are they single disk/partition backup or are they multiple disk backups. 

Are you backing up to something other than the NAS; and if so are there any problems with those backups? It is preferable to have multiple backup destinations; one should be kept off-site and/or Acronis Cloud.

The log snippet above is giving a reason why the file is being reported as corrupted!  There is a CRC mismatch or internal data structures mismatch for the T47_BakA.tibx backup chain.

This type of problem can have various different causes from memory issues, overclocking, cables, network or any activities happening on the destination drive that touches the backup files.

The first test I would recommend would be to create a duplicate task to a different destination such as an external USB drive and see if the issue repeats with that destination or not?  This will help eliminate the NAS and network as being involved if the same problem is seen with a USB drive.

If any overclocking is being used, then try dialling this back a notch and testing further.

IanL-S, Steve

My apologies for an oversight I need to correct:
The Acronis target backup drive is hosted by the NAS but is an external WD Mybook 1230
connected to the NAS via USB3
2.3 of 3.7 TB used

IanL-S-

• Yes backups run overnight
• The computer does not hibernate
• Synology NAS DS218+    
• The Recycle works reliably
  • Recycle Task Schedule- For all drives, runs daily limiting size to 500GB, delete old first

The local drive image backup that we are discussing that fails contains 2 partitions and runs daily at 12:30 am. Partition info- (note the Acronis archive below was created yesterday after I deleted the corrupted archive files and Acronis task, then recreated the task.)

A second Acronis backup task runs at 2:30 am, backing up folders-files (no partitions), from an internal NAS drive folder to the same USB backup drive above.

A third Acronis backup task runs from this same machine. It runs twice a month, on Sunday afternoons backing up my media files-folders from another internal NAS folder to the same WD target above

These run reliably and I am not backing up to any other devices

Steve-

There are other none-partition Acronis backups from this same source machine and Acronis instance (please see above). They are working reliably and the network and NAS work reliably throughout the day for all other processing.

Thanks for your help!

The Recycle works reliably

• Recycle Task Schedule- For all drives, runs daily limiting size to 500GB, delete old first

Please can you clarify the above statement?  What is this recycle task and how is it deleting old backup files first?  Does this mean it deletes backup files before ATI then runs to create new ones?

The local drive image backup that we are discussing that fails contains 2 partitions and runs daily at 12:30 am.

If the failing backup task is creating TIBX backup files, then these should never be deleted by any other process than by using either the ATI task automatic cleanup options (which will only delete after making a new backup) or else by using the 'Clean up versions' tool provided via the task menu.

Steve

The task info was in response to Ian who seemed familiar with the Synology NAS and when he asked " What cleanup rules do you use for the backup tasks?"  I assumed he was referring to the deleted files recycling but maybe I'm wrong. Hence, my response only concerned storage recycling which only impacts deleted files.

With this NAS, there is no default recycling like Windows provides. It is supported by using their Task Scheduler where you define a task to handle it. However, it works the same way as Windows recycling, and lets you set the frequency it should run and at what threshold of deleted files. It only purges files that have already been deleted and never affects any other files including tibx.

**

If you refer back to my initial post and the image of the ATI vault files on disk, for BAKA, there are eight files associated with the single backup. Is this normal behavior?

Thanks for clarifying the situation with recycling on the NAS - I also have a Synology DS215j NAS that I use to store my own backups on from multiple PC's and I don't have the recycle bin enabled on the NAS, so that any automatic cleanup actions to delete files (.tib or .tibx) result in the immediate deletion of the files on the NAS too, not moving them to the recycle bin and losing many GB of space on the NAS.  My NAS has 2 x 3TB WD Red NAS drives using RAID where using the recycle bin would take too much space!

For the file names shown in your File List image, these are correct for how .tibx works but the actual type of files would be shown in the Activity page of the backup task using the date / time stamps to match the individual files. This is because a file with a -00nn suffix can be a Full backup, Differential or an Incremental version chain.

The best method to try to determine why there is a problem with your T47_BakA task files would be to review the Backup Worker logs for the dates when this task has run (using the new MVP Assistant log viewer tool) to try to see where the error is introduced, if possible to determine?

Steve
Thanks for pointing out the Backup\Activity page which I had overlooked (it would be more helpful if it included the name of the file). Also, I only do full backups.

Back to Acronis making multiple files for a single full backup, I have another machine using Acronis which only backups the local C: partition using Full backup scheme. It never creates more than two files as below.
BakA.tibx
BakA-0024.tibx

The backup that fails had the same config and was working successfully but had created 7 backup files, failing as it was creating the eighth. Was it in an error state creating all these files when it was configured for a full backup?

The Backup Scheme settings control how many files you should be seeing.  For Disks & Partitions backup schemes, even with using Single Version scheme, there will always be 2 files once the task has run twice, as each task will have a 12kb metadata .tibx file plus the actual backup file.

I cannot tell you why your BakA task has / had created 7 backup files if it was set to 'Store no more than 1 recent version chains' for the backup scheme settings.  The logs would be the place to start looking, going back to the date / time when this task worked correctly and working forward.

I just noticed something I had overlooked in the Backup scheme screen

I had always use Single Version, where lower on the screen it indicates "Full backup version".

To try and resolve my  problems, some time ago I switched to Custom\Full which appeared to be the same as Single version as it also creates a full version.

I could never find any clear explanation in the docs between the two so I assumed they were the same but hoped that using Custom\Full  might somehow be more reliable.

However, today I see that when you select Custom\Full, a hyperlink is displayed to "Turn on automatic cleanup" which I had never selected.

When you select auto cleanup, the screen opens

So, without selecting anything other than the Full backup method, by default the backup was preserving 7 complete versions which I was unaware of and which explains all the files listed.

Technically, the screens for the both the Single and Custom Full options are not incorrect but their functions could be made more clear. The Single screen isn't too bad but perhaps the selection option could read "Single version with overwrite"

On the Custom screen, for each of the methods offered. it might append "with Version Control" and with the "Full" method changed to "Single" to be consistent. The link "Turn on automatic cleanup" could then be changed to "Version Control".

I think this would be much more understandable than only having the "automatic cleanup" making the distinction.

*

Using the MVP Log Viewer (great tool!) I checked all the logs and could find no further info.

For all backup tasks, I have changed all the backup schemes back to single file.