@momurda said in Backup solutions for Xenserver:
I use the Unitrends virtual appliance, and I am still loving my Unitrends setup. I'm not sure what you mean that XS is an afterthought for Unitrends. They are 1 out of about 3 vendors that support XS. Every problem(2) I have had with it in the last 8 months has been my fault. I am only backing up about 10TB and 40 or so vms though, perhaps your environment is a lot larger.
This is with 1 gbit interface between xs and ueb
It does do dedupe very well. I usually get about 40-60MB/s backup speed on full backups. Agent backups seem to be a bit slower.
With XS and unitrends you have to have your network interfaces setup correctly in XS and the Unitrends appliance, or all sorts of stuff doesn't work well, or at all. You also need to make sure you aren't trying to backup too many vms at once on the same SR, which can lead to tapdisk unpause errors, tapdisk timeouts and other bad things. This seems to be a limitation of XS 6.5, not sure if 7 is better.
My UEB has an interface on Network 0(mgmt. iface) of the XS host. And one other that is on another subnet so I can easily connect to the web server.
My environment is much smaller than yours, I have 12 or so VMs with about 6Tb of data. I do stagger the backups so that I do not go over the max connected VHDs to the appliance.
When you say that the network interface has to setup right, what do you mean? My appliance just uses the network gang I have (4x1Gb NICs) that is also used by all the other machines and as the management interface.
The main problem I've had was that it would error out during a backup, but that error would cause my SR to then fill up with snapshots, which then caused more errors in Unitrends since it would never detach the VHD or do any cleanup. When they looked through the logs they would then blame it on the Xenserver error and more or less stop supporting it at that point. The main error was SR_Backend_Failure_46, which I was able to trace back to the fact that the backups were failing.
Here's my setup:
2 Xenservers with an HP SAN for storage (connected via 8Gb/s fiber, in multipath mode). My backup target is a Dell server running Ubuntu with a single NFS share created. The appliance is using NFS as the protocol for backup.
The latest issue is that the appliance just sort of stops backing things up, as well as sending out emails about any failures or anything. And when I go to look at it I can see there are 4 or 5 machines queued and one or two just sitting there at some percentage and just not moving forward at all. I have not really bothered to call them on this, since I suspect they're going to just go have me re-install and re-impliment it again (for the 6 or 7th time in a year), I have just not had the time with our new school opening up.
At this point, even if they were to fix it, I am not sure that I would trust it. I mean PHD Virtual worked perfectly and reliably, but Unitrends 9 has only worked for me a total of about a month out of the last 12, and I don't have the time to spend hours and hours on the phone asking them to fix something that is clearly broken on their end.
To their credit, they have really gone out of their way to try and help me on this. So their support is top notch, I just wish their product was better.