Hello @fem 

Could you please share a screenshot for the backup job progress?

What is the Veeam backup server machine type (Virtual/Physical) and CPU/Memory?

How many Proxies do you have and its type (Virtual/Physical) and CPU/Memory?

What is the Veeam Repo type (SAN/NAS/Objects/Local Disk)?

What is network speed?

Besides the infos Moustafa asked, a good idea for VMware Backups is use HotAdd mode for the proxies. It should increase the backup speed if you are not using this mode yet. 

https://helpcenter.veeam.com/docs/backup/vsphere/virtual_appliance.html?ver=120

Hello @Moustafa_Hindawi 

Thanks for the comment.

There are over a hundred VMs in the backup job and only one failing backup.

The machine type is virtual with 4 CPU(s) and 24 GB Memory

We have 55 virtual proxies 8 CPU(s) and 16 GB Memory

The Veeam repository is SAN

Network speed 10.0 Gbps

To be a bit more clear, what i’m trying to work out is 

Can an estimate of how long it take to complete a backup be deducted from this (Hard disk 20 (3.5 TB) 1.9 TB read at 41 MB/s /CBT] - Duration 13:21:26) information? If so how can this be calculated?

How can i confirm if failed job is attempting a FULL of incremental backup?

Well, @fem 

I guess there are multiple factors need to be considered:

1st: Repo server compute resources CPU/RAM 

2nd: Disk format

3rd: Fast cloning registry

4th: error description

Being able to tell what kind of backup it is..full or incremental.. should be fairly straightforward. If it’s the 1st run of the job, it’ll be a full. Then, depending on your Backup Method (FFwd, Fwd, Reverse) and Retention Policy, you should be able to tell if the job run is still at incrementals or starting a new chain and performing a full. FFwd method will always do an incremental after the 1st run (full), because for this method to be configured you have to DESELECT synthetic or active full options.

There is a way to calculate backup speeds. This site here can help, but as @Moustafa_Hindawi states...more info is required.

The job is set to FULL on first run, incr after and synthetics weekly after.

The disk config was recently changed hence the reason i’m trying to find out if there is a way of confirming a failed job was doing a FULL or INCR.

Re calculate time

I was hoping based on info provided (Hard disk 20 (3.5 TB) 1.9 TB read at 41 MB/s sCBT] - Duration 13:21:26) the estimate backup time could be calculated without going into infrastructure bottleneck.

The repository you mentioned is a SAN - what kind of SAN?  How is it connected to the network - iSCSI, FC, NFS, SMB?  If it is a Windows VM connected to the SAN as a drive - are you using ReFS for the backup target?

Based on your screenshot of the job the Target is the bottleneck meaning the Repository.

Appreciate your replies, but they don’t exactly answer my question.

Can the backup duration be estimated just from this information (Hard disk 20 (3.5 TB) 1.9 TB read at 41 MB/s /CBT] - Duration 13:21:26)?

Assuming this is a FUUL backup.

We cannot answer your question without more information that was the reasoning for the questions from us.  Also you have the duration in your question so more clarity is required.

Hello @fem 

You can estimate the time if your backup performance is stable.

As i see from the graph, your backup performance is too low and unstable. The data read throughput (Green) and transfered data throughput (Red) are up and down. So, you can not estimate. 

Yeah...not quite able to do so without more info. You can see from the Calculator link I provided all the info needed to do the calculation. 

Thanks all, comments recieved encourages me to return with any further queries at a later date. I will work on utilizing the calculators, trying to work through it this moment.