Surebackup Issues

[GarethUK]

Hi,

I'm starting to use Surebackup and Virtual Labs. It all appears to work and even masqueraded connections from my main network are working. PowerNFS is a bit slow but using replicas seems to do the job. My target repos aren't ideal so we will be replacing them in the next few weeks.

However, I'm having some odd problem which I can't seem to explain which I don't think have been discussed in the forum previously.

When the SB job ping tester tries to ping a replica machine it does *NORMALLY* pass, however, only one of the tests per host will actually pass. In other words, I get a ping pass rate in the report of 12%. This also means it takes awhile to boot a machine as each ping test takes around 30 seconds to fail.

Pings from the Veeam server (or any other machine) to the masqueraded IP range works flawlessly. So, I'm a bit confused. I've had to route the masqueraded range across our network as my SB destination servers are located in our DR centre which isn't on the same network as the Veeam server.

Application checks seem to work but you obviously have to get the timing right.

My next problem isn't documented (from what I can find). If you backup a DC *without* using application aware snapshots the VM ethernet interface appears to be disabled (not connected/not connected at power on) as the machine turns on within the VLAB. The logging for the SB JOB seems to suggest it's configuring the DC but doesn't explain what this is. I agree that using a DC backup which isn't application aware for SB restores isn't ideal - I will address this. However, is this anticipated behaviour? To get around this I have manually connect the interface.


Regards,


Gareth

[foggy]

When the SB job ping tester tries to ping a replica machine it does *NORMALLY* pass, however, only one of the tests per host will actually pass. In other words, I get a ping pass rate in the report of 12%.

Gareth, could you please elaborate a bit on what particular tests fail?

[Vitaliy S.]

Hi Gareth,

My next problem isn't documented (from what I can find). If you backup a DC *without* using application aware snapshots the VM ethernet interface appears to be disabled (not connected/not connected at power on) as the machine turns on within the VLAB. The logging for the SB JOB seems to suggest it's configuring the DC but doesn't explain what this is. I agree that using a DC backup which isn't application aware for SB restores isn't ideal - I will address this. However, is this anticipated behaviour? To get around this I have manually connect the interface.

What adapter type do you use on this VM, E1000 or Vmxnet3? Is it a Windows 2008 R2 DC?

Thanks!

[GarethUK]

Thanks for the replies. The machines is a 2003 domain controller.

The test I'm referring to is the Surebackup ping test. Each test takes 27 seconds. The one or two that pass take 15 seconds.

The machine is running vm hardware version 4 on ESXI 5.0u3. The VMWARE tools for the host is out of date. The adapter is set as flexible.

Regards,

Gareth

[Vitaliy S.]

The machine is running vm hardware version 4 on ESXI 5.0u3. The VMWARE tools for the host is out of date. The adapter is set as flexible.

What happens if you change the adapter type to E1000? Will it be possible to check that? Also I would recommend reviewing VM log files to see why it cannot connect during registration on the ESXi host.

[foggy]

Do you mean that the ping test for some VMs in the job is successful and for others is not? Anyway, I also recommend to involve support here.