60114: Acronis Cyber Protect Cloud: incremental to a backup seeded via PDS fails with "Failed to create an incremental backup in backup archive"

use Google Translate

Last update: 31-05-2021

Symptoms

You are offering an Initial Seeding (Physical Data Shipping) service to your customer.

The customer successfully performs the first backup to a local location and then sends you the disk for upload with the Initial Seeding Tool.

A subsequent incremental backup fails.

One of the possible error messages is:
| error 0xa100fc: Stream name: 'computername-27584NEE-239A-4540-8E4E-2A172898D75A-CBB03110-9SC7-43B8-AEEA-1AF78EE58CEFAD.TIB'.
| line: 0x9fc3ee39511c2b90
| file: k:\3917\enterprise\mms\managers\archive\utils\uri_utils.cpp:533
| function: ArchiveManagement::AddStreamNameToErrForOnlineBackup
| account: XXXXX
| $module: disk_bundle_vsa64_3917
|
| error 0x1490003: Backup has failed.
| line: 0x1cd98aae889424f9
| file: k:\3917\enterprise\managers\gtob\util\impl\convert_batch_result.cpp:58
| function: Gtob::Backup::ConvertBatchResult
| $module: disk_bundle_vsa64_3917
|
| error 0x10424: Failed to prepare an operation.
| line: 0x145a191bfe9050c3
| file: k:\3917\processor\batch\batch.cpp:110
| function: Processor::BatchOperation::Execute
| $module: disk_bundle_vsa64_3917
|
| error 0x10449: Internal error: Failed to create the image builder.
| line: 0x51ea1282ec004ab
| file: k:\3917\processor\backup\stream.cpp:156
| function: DaProcessor::ImageStream::CreateBuilder
| $module: disk_bundle_vsa64_3917
|
| error 0x3f8: Failed to create an incremental backup in backup archive 'computername-27584NEE-239A-4540-8E4E-2A172898D75A-CBB03110-9SC7-43B8-AEEA-1AF78EE58CEFAD.TIB'. The base full backup cannot be found.
| line: 0x64a31b1b5afdaf3c
| file: k:\3917\processor\backup\imager.cpp:1655
| function: `anonymous-namespace'::OnlineStream::CreateJournalledStreamForWrite
| $module: disk_bundle_vsa64_3917

Cause

It is a known issue with the Physical Data Shipping process.

While the initial full backup is being shipped to your premises for upload, the customer runs the next backup as scheduled without waiting for the data to be uploaded. This might have the following consequences:

1. Incorrect values might be written the archive metadata (stored in an XML file).

2. A TIB file with zero size might be created in the storage and prevent backup task from reading the valid TIB file.

3. The IS_me command may fail with an error message stating that the respective archive with the same name is already uploaded and will be skipped.

Solution

Ensure that your customer has the latest build of the agent and you have the latest build of the IS tool.

How to avoid it:
Either disable the automatic schedule until the initial backup is uploaded or stop the Acronis Managed Machine Service until you confirm to the customer that the Seed is ready. In case the customer have the same backup plan applied to multiple machines, the second option is preferable.

How to fix it if this have already happened:
Ensure that the backup destination only has the following files at the start of the IS_me command:

<archive_name>.xml

Only these files once the IS_me is finished successfully:

<archive_name>.xml
<archive_name>.TIB (depending on the type of a backup, there will be an F or a D in front of the .TIB extension)
<archive_name>.TIB.jrnl
<archive_name>.TIB.meta (for file-level backup)

None of the files above should be of zero size.

 

More information

The product will be able to handle such scenarios in future versions.

Tags: