A really nice and important feature with Hyper-converged, a full recovery from snapshot a Virtual Machine shared via NFS.


What we need:

-access to SynetoOS web interface

-snapshot schedule set in place

-access to built-in ESXi VMware host


Scenario:

-we have a Virtual Machine, in our case a Vcenter with a corrupted database, where an update has gone wrong.

-we a have daily snapshot scheduled

-we will retrieve the snapshot before the update.



Long on to the SynetoOS web interface

IP is assigned by the customer.

default user & password: admin/admin


Identify the impacted Virtual Machine in our case:

DLNVCGB001 - a virtual Vcenter



  • Identify the snapshot taken before the upgrade:


                Go to Data management --> Folder/Virtual disks

                and select the dataset assigned to our impacted Virtual Machine, in our case Vcenter.

                go to snapshots

                select the snapshot taken before the upgrade and do

                ->Clone


                    Select the path and a name.

                -> Clone

                Important:

                Must use a different name

                We went with Vcenter_last



Now we should have a new dataset under tank/Vcenter_last

Next step is to share it via NFS:

Select the newly created dataset tank/vcenter_last

go to NFS share

-->allow new host

At the Hostname/Network

use the embedded ESX IP 172.16.254.1/32 to allow all the subnets

Select Read/Write and grant root access, Yes for Root.

 

Next step is to add the dataset as a new datastore in ESXi:

Long on to the VMware ESXi host using default IP and credentials: root/syneto123




Go to Storage tab-->New datastore-->Mount datastore:



Type in a generic name, we went again with /tank/vcenter_last

Important:

NFS Server's Hyper-converged IP:

172.16.254. -that's the internal vSwitch IP



Next and Finish.


Our new data store will show up under storage:


Next step is to test the snapshot dataset that we retrieved:


Go to Storage --> Register Virtual Machine and use the Vcenter_last recovered snapshot.

Start the newly registered Virtual Machine to test if the error has cleared.


If the error is no longer there then happy days, we retrieved right snapshot.


Next and final step is to overwrite the Vcenter_last on to our initial datastore vCenter:


-we are still on the ESX web client

-stop both Virtual Machines that are accessing Vcenter_last and Vcenter datastore, the original one

-go to Storage --> Datastore browser -->Vcenter and delete the containing data

details below


 -copy the data from Vcenter_last to Vcenter

goto to Vcenter_last datastore, select the folder and Copy

-use as a destination folder the old datastore Vcenter.


Copy process can take up to 2 hours depending on the size of the Virtual Machine.


Once this is finished we will have to re-register the Virtual Machine using the Vcenter datastore.


Start up the VM.