Hi,
When I try and run a backup job all guests fail to backup with following message:
Failed to create snapshot for {VM Name}, error -3948 ( vcb api exception)
Does anybody have any idea what this is because I can't see any configuration issues that would cause this and looking on the event log for the host it appears to successfully create the snapshots and then just take them off again? I am able to take snapshots of the guests with no issues.
Has this been working? Make sure name resolution is working both forward and reverse for both the clients and the host.
david
Has this been working? Make sure name resolution is working both forward and reverse for both the clients and the host.
david
Hi,
This is the first time I've tried it as I only just installed it today, so no it's never worked! I don't understand what you mean by name resolution, Data Recovery in vSphere shouldn't need to resolve names of machines as it backs up registered guests in the vServer inventory.
John
I know vSphere is very picky about correct name resolution. I have had issues with VCB not on vSphere and it was a name resolution problem.
david
I can see that may be an issue with VCB but there is nothing in the docs about name resolution of the guests being an issue, it should all happen through inventory and in any case the logs say the snapshots actually do work, and then the error is produced, and then the snapshots are deleted again - so there seems to be a system error here. I think maybe I should raise this with VMWare as it is a v1.0 product after all and who knows what bugs are lurking?
John
I've raised a support call. If they come back and it's name resolution I'll give you points .
In the meantime if anybody else has any experience with data recovery and an idea what this may be let me know.
Cheers,
John
As the poster COdlk has highlighted, this is indeed a known resolution issue.
The issue is that the 'VDR appliance' is unable to resolve the hostname of the ESX server on which the target (backup) VM resides.
If you add the ESX hostname to the /etc/hosts file of the appliance, you should find that the issue no longer occurs.
HTH
Cormac
Yes, this has worked - thanks.
I'll apply points respectively for COdlk giving me the right answer but me not believing them 😐 and you giving me a helpful nudge.
Cheers,
Jhn
I had this same issue when I set it up in a sandbox for testing.
Just for note for anyone else who stumbles upon this, your Data Recovery appliance must be able to properly DNS the vCenter Server, the ESX Host(s) and the VMs in order for it to work properly.
In my very small test environment, I don't have a DNS server, but editiing the HOSTS file on the Data Recovery appliance worked for me.
Thanks,
JV