File Level Restore not working - helper appliance timeout

[seanbrewer]

Hello-

I'm trying to restore a set of files from a CentOS VM. In version 5, it appears the only way I can do this is to spin up the hehper appliance on a host server. This isn't a huge issue, but my problem begins here.

After setting up the initial restore params in the wizard, I wait for the helper appliance to spin up and present the files available for download. Unfortunately, that never happens. Instead, I get an error and this message why it wouldn't start:


10/22/2010 6:42:01 AM Multi-OS File Level Restore started
10/22/2010 6:42:42 AM Mounting vPower NFS datastore to host "(ip address)"
10/22/2010 6:42:43 AM Mounting VM disks
10/22/2010 6:42:43 AM Compiling FLR appliance configuration
10/22/2010 6:42:51 AM Registering FLR appliance "VeeamFLR_(ip address) " on host "(ip address)"
10/22/2010 6:42:54 AM Powering on FLR appliance
10/22/2010 6:52:58 AM Fail FLR appliance IP address is "{0}"
Timed out waiting for FLR helper appliance to start.
10/22/2010 6:53:01 AM Fail Unknown failure
Timed out waiting for FLR helper appliance to start.

Any ideas?

Thanks.

[Vitaliy S.]

Hello Sean,

It seems like the appliance wasn't able to start within 10 minutes, you may try to set this registry key on the backup server to extend the timeout, if it doens't help please send us all the logs to support@veeam.com for further troubleshooting.

The registry key (in seconds):

Code: Select all

DWORD FLRVmToolsWaitingTime

Thanks!

[seanbrewer]

Thanks for the quick reply!

I've tried that fix, but get the same result- fail-timeout.

Any other ideas?

[Vitaliy S.]

Sean, please contact our support team for assistance directly as we need to check out your log files to find the resolution.

[Gostev]

I am guessing Virtual Appliance starts on the different network from Veeam Backup server, so our server simply cannot reach it. Please verify the settings you specified on the last step of the wizard (where you chose ESX host to run virtual appliance on).

Increasing timeout is quite pointless, as normally appliance boots in like 10 seconds (so default 10 mins is already much bigger timeout than needed).

[albertwt]

so by default is it 600 decimal right ?

@Sean, make sure the vSwitch network label is the one that is the same as your production to access DHCP server.

[Vitaliy S.]

so by default is it 600 decimal right ?

Yes, that's right.

[fate]

Sorry to bump such an old issue but with VEEAM b&r 9 does not honor the "FLRVmToolsWaitingTime" regkey anymore.

Any idea how to fix?

My VM takes around 14min to mount the partition from the DataDomain

[foggy]

This registry value defines timeout for virtual appliance guest IP detection, while you're talking about slow mounting of the disk from the backup, right? What error do you get? Have you contacted support already?

[fate]

I'm in the middle of creating a ticket. Case #01805967

Yes mounting is slow... But that value fixed it in version 8... It just needs to wait longer before the Helper instance responds.

Error:
23.05.2016 13:41:17 Error Failed to retrieve FLR appliance IP address Error: Timed out waiting for FLR helper appliance to respond. Possible causes:
1. Appliance is configured to run on different network than backup server.
2. Appliance is configured to obtain IP address automatically, but DHCP server is not available.

EDIT: to clarify this only happends with large VMs (2,8TB)
Small VMs 50-250gb work perfectly fine.

[foggy]

Well, I've checked with QA and actually we now have another value that is responsible for appliance IP detection: FLRApplianceIPWaitingTime. The previous value affects the mount operation itself.

[bconn-rverdam]

Do we need to restart the backup service for this setting flrapplianceipwaitingtime to become active?

[bconn-rverdam]

Feature request:
It would be great that these kinds of advanced parameters (timeouts, log locations, log rotation etc.) would be configurable via the GUI and activated via the GUI (maybe via some advanced mode or something).

[foggy]

Yes, restarting the backup service is required for the value to take effect.

[skrause]

I had similar errors with FLR that were caused by the machine running the console being blocked by our firewalls. Might want to check your firewall logs and make sure the traffic isn't being blocked.