Backup virtual machine with large volume (raw device) stuck

[rejaine]

Hi,

I have a virtual windows machine on a Vmware 6.7 cluster, with this setup:

OS Guest: Windows 10 Workstation

Hard disk 1 120 GB | SCSI(0:0) --> VM datastore (SSD)
Hard disk 2 1024 GB | SCSI(0:1) --> raw map device (storage 3par LUN SATA)
Hard disk 3 1024 GB | SCSI(0:2) --> raw map device (storage 3par LUN SATA
Hard disk 4 1024 GB | SCSI(0:3) --> raw map device (storage 3par LUN SATA)
Hard disk 5 1024 GB | SCSI(0:4) --> raw map device (storage 3par LUN SATA)

The hard disks 2, 3, 4 and 5 correspond to a single volume E: on the Windows machine (total GuestOS size = 4TB, formatted with NTFS)

I need to backup this volume E: , so I doing a this:

Veeam Backup & Repication Enterprise Plus Version 10

Backup Type: Windows Workstation
Backup Mode: Volume Level Backup
Managed by: Agent (I tring by server too, but not success)
Source: E:\
Destination: Backup repoosity (StoreOnce 4400 CIFS NAS SHARE, with dedup)

The backup not finish after several hours stuck in 99%) and displays some strange statistics such as:

Some statistics after 22 backup hours :

Processing rate: 35 MB/s (average)
Status: 88% (stuck)
Processed: 45,7 TB (99%) ---> why this???
Read: 3.1TB
Transfered: 41.4 TB (0,1x) ---> why this???

I tried changing it to "Managed by : backup server" or using the "file level backup " backup method (very slow.. took more than 4 days to complete)

I don't know what else to do to get this volume backed up.

Any tips?

[PetrM]

Hello,

Please provide a support case ID as requested when you create a new topic about a technical issue. As explained there, we cannot troubleshoot technical issues effectively over the forum posts. All posts about technical issues without case ID will be eventually deleted by the forum moderator.

Anyway, I'm curious to know why you don't add vCenter/ESXi to the Backup Infrastructure to use vSphere backup job instead of the agent one?

Thanks!

[rejaine]

Hi,
Case #05786788
I have a virtual windows machine on a Vmware 6.7 cluster, with this setup:

OS Guest: Windows 10 Workstation

Hard disk 1 120 GB | SCSI(0:0) --> VM datastore (SSD)
Hard disk 2 1024 GB | SCSI(0:1) --> raw map device (storage 3par LUN SATA)
Hard disk 3 1024 GB | SCSI(0:2) --> raw map device (storage 3par LUN SATA
Hard disk 4 1024 GB | SCSI(0:3) --> raw map device (storage 3par LUN SATA)
Hard disk 5 1024 GB | SCSI(0:4) --> raw map device (storage 3par LUN SATA)

The hard disks 2, 3, 4 and 5 correspond to a single volume E: on the Windows machine (total GuestOS size = 4TB, formatted with NTFS)

I need to backup this volume E: , so I doing a this:

Veeam Backup & Repication Enterprise Plus Version 10

Backup Type: Windows Workstation
Backup Mode: Volume Level Backup
Managed by: Agent (I tring by server too, but not success)
Source: E:\
Destination: Backup repoosity (StoreOnce 4400 CIFS NAS SHARE, with dedup)

The backup not finish after several hours stuck in 99%) and displays some strange statistics such as:

Some statistics after 22 backup hours :

Processing rate: 35 MB/s (average)
Status: 88% (stuck)
Processed: 45,7 TB (99%) ---> why this???
Read: 3.1TB
Transfered: 41.4 TB (0,1x) ---> why this???

I tried changing it to "Managed by : backup server" or using the "file level backup " backup method (very slow.. took more than 4 days to complete)

I don't know what else to do to get this volume backed up.

Any tips?

[rejaine]

First time on forum, sorry...

Where I put the number of my Case #05786788 (on subject or body?)


The problem is that the license is registered in the email of another account (my boss) and my product support account only shows open/free licenses for me, even so I opened the ticket..

[jamcool]

If using Windows agent and volume level backup it is most likely within the server's operating system. Assuming CPU or Memory not MAX out:

1. on the server, open a command prompt and run VSSADMIN List writers. Make sure they all say stable and no error. If one is not, Google it and it should tell you the service that needs to be restarted. If that does not work, may need a reboot to free up that VSS writer. Common ones that are not stable (or could be missing) are System Writer and WMI Writer.
2. check to make sure not out of disk space as the agent will use VSS to create a shadow copy of the disk which requires some free disk space.
3. go to properties of drive letter, select Shadow Copies tab and make sure can do a create now on each volume. Be sure to delete them. Also, if you see any existing ones, may want to delete them unless know why they are there.
4. While there, go to Settings for each volume and change from Use Limit to No Limit.
5. Maybe check Eventlog (system and application) to see if anything there.

In the end a reboot is always good. . I had supported TSM for almost 18 years on 1500 servers all with clients, you pick up a few things.