Strange behaviour

[pizang]

I am backing up a virtual machine that has 1240 GB of disk space in total. The backup started on 7:16 AM (full, fresh one). Before 1 PM I noticed that backup is showing 99% of completion. Average transfer rate was 93MB/s. It finished at 6PM and average transfer dropped to 32MB/s.

The question is what Veeam was doing for 5 hours? As it is a full backup it couldn't calculate some incremental data. I have seen on my storage that it is still transferring something as CPU was busy. What was that?

[Vitaliy S.]

Hello pizang,

Could you please tell us what is your target storage? Also It'll be great to see the logs from Help->Support Information for this jobs (send the log to support@veeam.com), so we could see your transfer rate and tell you exactly what is the reason for such a drop down.

[pizang]

Hello,

I will send logs in a minute. I use QNAP TS634 with 6x1,5TB disks in RAID5. Disks are exported through SMB service.

[Vitaliy S.]

Hello Miroslaw,

After investigating the logs and keeping in mind what's your storage and how everything is connected, we believe that this issue most likely is related to SAMBA issue, cause when reading/writing small blocks of data from a storage using CIFS (SMB) share, it causes the delays that we see in your logs when the backup job backups VMware logs and other small files. So basically, the problem is related to:

1. Backup index rebuild taking a lot of time. This operation is reading and writing small chunks of data describing the content of the backup.
2. Performance degradation when writing actual backup data to the storage. Actually data write speed remains nearly the same but due to
the way we calculate speed of backup for user it looks like a degradation.

[Gostev]

Does your storage has an option of using NFS instead of SAMBA/CIFS? In that case, connect it to the Linux server and add this server as a target in Veeam Backup job settings, this should help.