Veeam FLR

[mdking]

Hi All,

I just wanted to confirm something regarding to the use of FLR in Veeam.

Background information:
Veeam Server is a VM
The default proxy is being used
File Server is on a VM currently using 4.4TB.
VSphere 6
Veeam 8 latest update

We seem to have a problem when using Veeam FLR and restoring files from the file server. When we use FLR and mount the file server the Veeam server freezes.
I have spoken to Veeam support, they mentioned that I should use instant recovery. (Which works fine). But its quite a long and laborious task.

I use the FLR function on other VM's which are much smaller. These are fine currently.

Any thoughts?
Thanks

Matt

[Vitaliy S.]

Hi Matt,

Are using built-in FLR engine or other OS restore wizard? Also what's your case ID with Veeam?

Thanks!

[mdking]

Hi,

I have tried both.

There has been a few Case IDs:
01063539

01080092
00898188

[Vitaliy S.]

Hmm...what is your backup target device?

[mdking]

Its a Qnap ts1279u-rp

[Vitaliy S.]

Trying to find out the details in your tickets on how your server freezes during A. built-in FLR and B. Other OS. Can you please provide a bit more info? Is it the restore process that is slow or mounting data to the backup server completely disables your backup server?

[mdking]

Hi,

I have tried the Built in FLR and the Other OS. If I choose a small VM (100GB) the FLR amounts fine.
When I try to use the FLR on the File server (4.4TB), it starts the process, then the Veeam backup server freezes.

[Vitaliy S.]

File-level restore performance depends on the repository hardware, not sure how many random IOPs your device can provide, but it seems like there could be indeed some issues with FLR process. Try to do indexing for this server and then use EM to restore files as per this post from Tom.

[Buh95]

Hi,

I recently have some problems restoring a lot of files (60k for 270GB) using FLR. The restore had as target the veeam server itself (physical big server).
The FLR restore was starting fine and failed at 20% with this message : Failed to open management RPC connection proxy service. Port: '6160'

We finaly found that it was due to a lack of TCP client port during the restoration. We tuned the TCP parameters creating those entry :
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\Tcpip\Parameters:
TcpTimedWaitDelay (DWORD) = 30
MaxUserPort (DWORD) = 65534

That's it... not exacly your problem but I would check this on your veeam server if you try to launch big FLR restore.

[mdking]

Hi,

Strangely we have seen this message before, but not on the File Server (on another VM).
Im still not sure if the freezing could be due to this?


Thanks for the information

[Vitaliy S.]

Yes, it does not relate to your problem, but still a good tip to know.

[mdking]

File-level restore performance depends on the repository hardware, not sure how many random IOPs your device can provide, but it seems like there could be indeed some issues with FLR process. Try to do indexing for this server and then use EM to restore files as per this post from Tom.

Hi,

I shall look at this and get back to you. Is there any information on the QNAP that would be of some help?

Thanks for your help!

[Vitaliy S.]

Don't know if it can be feasible in your case, but try to copy a VBK you're having trouble with to another storage and check the performance of the FLR wizard again.

[EBoucq]

Hi,

Any idea why I get a "Failed to connect to Installer Service. Failed to open Management RPC connection to Proxy service. Pört 6160" while trying to restore thousands of files on a local Veeam server backup drive ?
Thanks !!

Eric

[EBoucq]

Why do you have temporary flat vmdk files in windows temp folder when you try to restore guest OS files ?
How much space do you need when you want to restore guest files on Veeam server ?

Thank you

[foggy]

Those are not actual VMDK files, but stubs required to present disks during FLR. They do not consume space.

Please review the thread above for the hint regarding the error message you see.

[cffit]

This worked for me. I was getting a similar error copying a folder that was 18GB in size. Not exactly this error, but errors about RPC and connections when I was about 2/3 done copying. When I made the changes to the registry I was able to copy from backup fully. Thanks for posting this.

Hi,

I recently have some problems restoring a lot of files (60k for 270GB) using FLR. The restore had as target the veeam server itself (physical big server).
The FLR restore was starting fine and failed at 20% with this message : Failed to open management RPC connection proxy service. Port: '6160'

We finaly found that it was due to a lack of TCP client port during the restoration. We tuned the TCP parameters creating those entry :
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\services\Tcpip\Parameters:
TcpTimedWaitDelay (DWORD) = 30
MaxUserPort (DWORD) = 65534

That's it... not exacly your problem but I would check this on your veeam server if you try to launch big FLR restore.

[EBoucq]

Changing the registry to increase the number of TCP client ports solved my problem.
Thanks !

[wijngaarden.m]

Under C:\Windows\Temp folder we've found .flat files, with in the name referring to Veeam FLR.

Can anyone explain the functions of these files, and if we can delete those to clear up disk space?

[pshute]

I think they're used by restores and instant recoveries, and can be deleted if there are none running. I suspect they're left there if the service is restarted, or something like that, but I'm not sure. Discussed here:
https://forums.veeam.com/post276590.html