Sudden increase in synthetic creation time

[aw123]

First things, I'm far from an expert when it comes to Veeam!

We currently run Veeam in combination with Symantec Backup Exec, to backup 32 VM machines. Up until 2 months ago, we had been using this for quite some time without issue. We run incremental backups Mon - Thurs and a Synthetic full every Friday. After these issues started we also added in for an active full to run on a monthly basis.

The problem is, previously the Veeam synthetic full was taking less than 10 hours, including 9 hours for the "Synthetic full backup created successfully" process. For an unknown reason, this suddenly changed to taking almost 20 hours, including just over 19 hours for the "Synthetic full backup created successfully" process. It has done this every week for 2 months. We have deleted and recreated the backup job, deleted the folders on the backup repository and added in the active full - all to no avail. We backup directly to a physically connected SAN storage, this is then copied to tape.

The problem is due to time constraints the added time duration is impacting the tape backup, as previously the Veeam backup would finish 4+ hours prior to the tape backup beginning. At the moment backup to tape is starting prior to Veeam completing which causes a knock on impact for our Monday nights backup also as it has to backup additional data.

Statistics for these two backups are as follows:

previous quick job

Summary
Duration: 9:49:43
Processing Rate: 65 MB/s
Bottleneck: Source

Data
Processed: 2.9TB
Read: 96.3GB
Transferred: 50.0GB (1.9x)

current slow job

Summary
Duration: 19:45:09
Processing Rate: 66 MB/s
Bottleneck: Source

Data
Processed: 3.0TB
Read: 108.7 GB
Transferred: 51.6GB (2.1x)

One thing I have also noticed is that despite the retention period being set to 5 previous restore points, I currently have 7 files in total - 2 x .vbk & 5 x .vib

Any help to resolve this would be appreciated. Thanks in advance

[foggy]

Were there any changes performed to your target storage at the time when performance started to degrade? Synthetic activity is very I/O intensive and requires random I/O, seems everything is fine with the backup stage itself, however, synthetic full became slower, which most likely tells about issues with the target storage.

[Gostev]

Please also include support case ID according to the forum rules provided when you click New Topic, otherwise the topic will get deleted during the periodic cleanup.

[aw123]

Ive logged a support ticket (ref:01728830)

In response to the questions asked, we have made no changes whatsoever to the storage.

[foggy]

Let's see what our engineers could conclude after reviewing the log files.

One thing I have also noticed is that despite the retention period being set to 5 previous restore points, I currently have 7 files in total - 2 x .vbk & 5 x .vib

To explain this part, please review how retention works for forward incremental mode.