New to VB&R 7.0

[bubbawny]

Hi folks:

I represent the IT dept of a new client for VEEAM. We just purchased and installed VEEAM B&R 7.0. I have a couple of questions that might be considered fundamental and I'd appreciate any thoughts.

We have a tremendously simple environment. We have (1) Dell PowerEdge R720, 32GB RAM, 2.6TB DAS (SATA), with (4) GB NICS (Unteamed because we don't have a managed switch...yet). We are running VMWARE ESXI 5.5 Essentials with (3) VMs.

VM1 is a W2012 Domain Controller / File Share
VM2 is a W2012 RDS server hosting most of our apps.
VM3 is a W2012 as a management/utility server. it's running Sharepoint Foundation 2013, Symantec Endpoint Protection, and VEEAM.

We purchased a SYNOLOGY DS412+ running DSM 5.0. We have configured a few iSCSI LUNs.

1 LUN is targeted as a secondary datastore via iSCSI from our VMWARE host. It's really not in use, outside of having hosted a copy of the folders from datastore1 (manual copies using SSH and VMKFSTOOLS).

1 LUN is attached to our DC as Drive E: (iSCSI) and is a volume dedicated to Windows 2012 Built-in Backups.

1 LUN is attached to the Sharepoint VM as Drive E: for being a repository for VEEAM Backups.

With all that said, here's my question:

Is there a way I can use VEEAM to take raw copies of the DATASTORE1 folders and move them to the secondary iSCSI store on the host?

My thinking is this:

Our VEEAM backup server is a VM. Not really good if our host gets trashed, but I don't have any other physical servers to put VEEAM on.

I really don't care of the ESXi host gets trashed and needs to be rebuilt. It's a really simple thing.

But in order to recovery in that scenario, i'd have to install VEEAM some where to recover the VMDK and other files. ALong with rebuilding a new ESXi Host, presuming it was trashed for some reason, I'd have to figure out a way to install VEEAM, connect to the iSCSI volume dedicated to the VEEAM Backup files (DRIVE E: on our sharepoint VM), only to finally restore the VMs and move them to a newly created ESXi host.

What i'm thinking of / desiring, is a way to make a scheduled job to copy the folders in the ESXi datastore to the iSCSI datastore as ready made folders/VMDK files/config files for near instantaneous mounting on a separate / new host.

I've managed more enterprise virtualization environments. Ours is not. We have a low end Dell, a low-end "SAN" (DS412+) and are a small business who does not have the $ to invest into multiple hosts/with a real shared SAN to be able to use VMOTION and other goodies to make for quick recovery.

In my reality, if the PowerEdge R720 gets hosed, it'll be an easy thing to repair/replace. In that scenario, I'd like to be able to simply reconnect to the iSCSI volume on the DS412+, and bring up the three VMs. From there, I could restore any files that may need to be updated since the last backup. Mostly, though, by having a standby set of VMDKs, I could quickly mount my management VM and do any restores using VEEAM without having to build a new Windows Server VM/Install VEEAM, Recovery Manually.

So that's my question in a long winded way:

How can I use VEEAM to help me in a sense run a job that copies the (3) VMs to the iSCSI volume hooked up to the ESXi Host.

At present, I have a job that is running in VEEAM to backup all three VMs via the Sharepoint Server connected to it's own iSCSI volume. It's working. I"m just thinking with one host, I need to be able to bring up / recover the VMDK files as fast as possible.

In the future, I hope to get another ESXi server going, but this company will likley never invest the $ into a real SAN and we'll likely continue to use DAS.

(In a previous job, I had a dedicatd HP/LEFTHAND iSCSI SAN with multiple ESXi HOSTS that made recovery nearly instantaneous. I just don't have that equipment here. THe PowerEdge R720 is a basic (slow) server with SATA storage, and the DS412+ populated with (4) SATA drives is just to slow and SOHO to use it as the foundation for HOST Storage.

I've got limited options, but would feel much more comfortable at night if I had READY copies of the VMDKs standing by on another datastore ready to be activated.

I know I"m a noobie still with VEEAM. Just hoping it has an optoin that can help here.

Our VMs

[ashman70]

You don't mention what the datastore for your Veeam backups is. You can install Veeam on any Windows 7/8 64bit host, it doesn't have to be installed on a server. If you could get a workstation with a 64bit version of Windows on it, then you could install Veeam on that, I would then suggest creating space on the Synology and making that the datastore for your Veeam backups. This way if the ESXi host goes down, Veeam is ok on the workstation, your backups are ok on the Synology NAS and all you have to do is possible reinstall ESXi on the server and restore from your backups. Just my 2 cents.

[foggy]

There's also a standalone extract.exe utility, which comes with every Veeam B&R installation and being copied to the backup storage along with backups allows to perform restore even without the need to install Veeam B&R first.

How can I use VEEAM to help me in a sense run a job that copies the (3) VMs to the iSCSI volume hooked up to the ESXi Host.

You can use File Copy jobs for that, however keep in mind they are not incremental.

[veremin]

What i'm thinking of / desiring, is a way to make a scheduled job to copy the folders in the ESXi datastore to the iSCSI datastore as ready made folders/VMDK files/config files for near instantaneous mounting on a separate / new host.

I don't think that using file copy job is a feasible idea, since it can't capture incremental changes, transferring full files each cycle. Let alone absence of retention, quiescence and many other settings that backup job has. So, backup jobs are still preferable. If you want to have fastest restores in case of production lost/recovery, then, utilize Instant VM Recovery that allows running VMs directly from backup files.

Thanks.

[bubbawny]

Thank you.

If it wasn't clear, the VEEAM VM I have has an iscsi LUN targetted on the DS412+. That lun would be able to be targetted by a different machine if the host went down. My concern was how to get the VMDK out of the backup file and restore it to a new host. I will look into the file extract. If I can learn how to extract the VMDK, I'll sleep better.

I think i will experiment today. I will try the file copy job.

I also like the idea of putting VEEAM on workstation. The issue is that they shut all their PCs off each night. I'll have to have my friend agree to keep one of the PCs on. The other issue is that the network connections from workstations to the server is 100MB (on the client side) as he's got an old network. I've been asking him to put in Gig switches, but he's not relented.

Thanks for the ideas and info, folks. I realize i'm new still to VEEAM. I'm enjoying the software so far. It's working with one issue (temporary snapshots for one of the VMs is not deleting). But all in all, i'm please so far.

[foggy]

If it wasn't clear, the VEEAM VM I have has an iscsi LUN targetted on the DS412+. That lun would be able to be targetted by a different machine if the host went down. My concern was how to get the VMDK out of the backup file and restore it to a new host. I will look into the file extract. If I can learn how to extract the VMDK, I'll sleep better.

Yes, it was clear. Keep us updated on your testing and feel free to ask any questions, if you need some further clarification.

[veremin]

I'm enjoying the software so far. It's working with one issue (temporary snapshots for one of the VMs is not deleting).

In the process of creating/deleting snapshots VB&R plays only "requester" role, sending corresponding instructions to VMware to create/consolidate snapshot. Thus, the problem with orphaned snapshots is typically related to VMware rather than to Veeam side.

In general, the issue shows up when the VM original disk is still locked by another process at the time the remove snapshot command is received by the host. The unique identifier of snapshot is removed from the VMware database before the actual snapshot is deleted. As the result, the snapshot exists, but is no longer shown in Snapshot Manager.

Another reason might be high disk I/O latency at the time of consolidation process; it might be worth reviewing this article for better understanding.

In order to remove orphaned snapshots, you can create another snapshot, then perform a “delete all” operation in Snapshot Manager. All snapshots, including the ones not listed in snapshot manager will be consolidated. Also, you can use Veeam ONE to be informed about orphaned snapshots' presence.

Thanks.