Mount failed on Linux file level restore

[saxie]

Hi All

Running B&R v8, vmware, esxi 5.5

I have a Linux VM with a 6Tb virtual disk. when trying to do a file level restore I get the error "cannot mount". I'm thinking this is because the FLR appliance is deployed as h/w v4 when the wizard starts and this does not support 6Tb size volumes? I've tried to modify the h/w version of the LiveCD template in Veeam but then I just get a error about my version of ESXi does not support powering on a VM of newer h/w versions. (strange, since I'm running ESXi 5.5 on the host where the appliance gets deployed.) Any ideas?

[Vitaliy S.]

Hi Werner,

Upgrading hardware version of our appliance is fully supported, I've seen many times when our users did the same thing as you. To what version of hardware did you upgrade your VM appliance?

Thanks!

[JeffCT]

B&R 8, esxi 5.1, emc vnxe

This is a problem for us as well (was an issue in b&r 7 also, was hoping it would be fixed in on our Cent 7 XFS vm's..... 6.5 with ext4 is fine and file-level restores are perfect. There is something about the new cent 7 version and xfs that doesn't work at the file-level. This is a huge problem for us and may lead us to move away from the product since we rely on these types of restores more often than not. I spoke with some of the Veeam folks on the phone and they know it's an issue but couldn't confirm if it was fixed in 8 or would be in a subsequent patch of 8.

Luckily this is in our dev environment and we can work around it a bit, but looking to move to this in production and it's probably not going to happen.

Can you confirm your Linux version and file system type?

[tsightler]

Support for RHEL7/CentOS7 + XFS is coming. There was already test code available for the v8 RC but it was too late for that to make the QC effort for the v8 GA. I'm personally hopeful that this will make the first patch but until that point I'd strongly suggest that each customer needing this open a support case to make their voices heard.

[saxie]

Case # 00677618

[saxie]

VitalityS, thanks for the repsonse. I've powered off the appliacne and upgraded it to the latest h/w (v10) but then, of course, the wizard has 'lost connection' to the appliance and restarting it simply deploys it with the original h/w v4. If I modify the template it deploys as h/w v10 but then it won't power on - it errors out - see my first post..

I'm trying to restore from Cent 6 VM...

[JeffCT]

>>>"Support for RHEL7/CentOS7 + XFS is coming. There was already test code available for the v8 RC but it was too late for that to make the QC effort for the v8 GA. I'm personally hopeful that this will make the first patch but until that point I'd >>> strongly suggest that each customer needing this open a support case to make their voices heard."

Tom, Case #00657483

This was their response:

"Hello Jeff,

Unfortunately, we won't know if CentOS 7 will be supported for file-level restores in version 8 until the official release. Even if it's not supported in 8.0, support would likely be included in a future patch for version 8.

Michael Franz
Veeam Support"


Basically, no workie....

Is there any way to test this beforehand or now? It's honestly a critical issue for us since we are about to launch a new version of our product into production with all Cent 7 vm's and have a number recently deployed in dev. We want to purchase Veeam 8 for our new deployment, but I'm looking at alternatives since it doesn't work. There are ways around it, yes, but we need quick access at the file-level, not restoring the entire thing, spinning it up, manually finding what we need, etc.

Thanks,
Jeff

[saxie]

Update...

When I tried to change the FLR template's H/W version to 10 I made a typo... so that is why it wouldn't power on. I though it was easy to spell '10' ... Anyway, now that the template is set for v10 it at least powers up but I still get the error when mounting the vmdk: ext3fs, cannot mount

[tsightler]

Jeff, I'll PM you. I don't want to derail the OP topic as it appears his issue has nothing to do with XFS.

[JeffCT]

Thanks, sorry for the off-topicness of this....