New proxy causing backup jobs to fail

[Kynaeus2]

Hi everyone,

Background: ESXi x3 at 5.1, vSphere at 5.1, B&R at v6.5. iSCSI SAN for the VMs, backup destination is a NAS

So, in a thread on reddit.com/r/sysadmin I was asking about how to do more complex scheduling to get around long job times for active fulls on the weekend, someone suggested I add a proxy to decrease processing time. Fine, the steps look very easy. I deployed a blank Server 2008r2 standard VM yesterday and added the backup proxy in Veeam successfully, tried to run a job, and they fail within a moment or so if they choose the new proxy.

Backup failure error: http://imgur.com/DBZzWuz

I adjusted the proxy settings so it matches the original one created during installation (the Transport mode was automatic, I changed it to Virtual Appliance to match), same failure message. Datastore access was still automatic, I changed it to manual to double check what ones it could access and discovered the list of datastores it can access is not correct, it appears to only have access to the ESXi hosts' local disks, which of course have nothing stored on them. (Reference: http://imgur.com/OImCuyv) I assume this is the real problem, it can't find any of the VMs to backup and fails

As I said, the instructions for adding proxies is very simple, which leads me to believe I've missed something obvious that is still eluding me. Any ideas would be appreciated

[alanbolte]

If you can't see the shared storage in that list, it might be a permissions issue.

Also, if you are backing up to a network share on e.g. a NAS, and you haven't specified a gateway in the repository settings, make sure the proxy has access to the share.

[Kynaeus2]

Sigh.

I spent a few hours looking around AD, Veeam, and the NAS admin panel looking for how access was being provided or restricted only to come up dry. The NAS allows for the capability but was set to the widest reaching most accessible settings, nothing restricted. I finally figured out what was allowing the Veeam server alone to access the NAS and its shares: a modified hosts file.

-.-

The NAS is now registering itself in DNS and proper access controls are in place and documented, the new proxy just finished successfully backing up an incremental job as a test. Now that I have two, I can schedule two concurrent jobs, right?

[veremin]

It's not about number of concurrent jobs, but rather about number of tasks (virtual disks) that can be processed by existing proxy servers concurrently. This number is defined by these settings configured on proxy servers.