Preparing the Destination C-mode system

Make sure SSLv3 is enabled:

system service web show

It was suggested to enable FIPS however the Unified Manager couldn’t then monitor the system.

Check that the Cluster Management lif has the right firewall-policy applied to it and that the policy allows http and https:

net int show -vserver netapp02 -lif cluster_mgmt -fields firewall-policy system service firewall policy show -policy mgmt

Make sure there are intercluster lifs in place mapped to the 10g ports so the snapmirror traffic goes over the fast interfaces. These are dedicated to snapmirror traffic  and do not failover upon head failure which is why ther should be one per node in a cluster. You can use already in place Snapmirror intercluster interfaces bit if there are no intercluster interfaces then create them.

If VLAN interfaces and non default IPspaces are used then the the snapmirror lifs needs to use a VLAN interface that’s in the default IPspace. For example:

netapp04::> network interface create -vserver netapp04 -lif nas40a_sm -role intercluster -home-node nas40a -home-port e0c-29 -address 172.29.1.30 -netmask 255.255.0.0 -status-admin up
Warning: The configured failover-group includes no valid failover targets for the LIF's failover-policy. 
To view the failover targets for a LIF run "network interface show -failover".

In the above example the failover warning is because netapp04 has a single 10g network connection. Even when using VLAN interfaces on top of an interface group this warning will appear.

Once created or if they are already in place, check the routing for the cluster SVM to the source filer and add a route for the source filer/vfiler network if necessary. Then ping test from all intercluster lifs:

netapp02::> net route show -vserver netapp02
(network route show)
Vserver             Destination     Gateway         Metric
------------------- --------------- --------------- ------
netapp02
                      0.0.0.0/0       172.20.1.1      20
                      172.29.0.0/16   10.7.17.1       20
  2 entries were displayed.
netapp02::> net ping -lif nas30a_sm -vserver netapp02 -destination nas20
  (network ping)
no answer from nas20
netapp02::> net route add -vserver netapp02 -destination 192.168.16.0/22 -gateway 10.7.17.1 -metric 20
(network route add)
netapp02::> net ping -lif nas30a_sm -vserver netapp02 -destination nas20
  (network ping)
  nas20 is alive
netapp02::> net ping -lif nas30b_sm -vserver netapp02 -destination nas20
  (network ping)
  nas20 is alive
netapp02::> net ping -lif nas39a_sm -vserver netapp02 -destination nas20
  (network ping)
  nas20 is alive
netapp02::> net ping -lif nas39b_sm -vserver netapp02 -destination nas20
  (network ping)
  nas20 is alive

Preparing the Source 7-mode system

We set the snapmirror.access to all because if the cluster’s SVM is added manually to the snapmirror.allow file then an internal snapmirror script will over write it and break access.

options snapmirror.access all

Make sure https admin is possible, enable SSL, disable SSLv2&3 and enable TLS:

options httpd.admin.ssl.enable on
options ssl.enable on
options ssl.v2.enable off
options ssl.v3.enable off
options tls.enable on

Make sure Snapmirror is possible on all interfaces:

options interface.blocked.snapmirror ""

Make sure the i2p, read reallocate, and nvfail volume options are off:

vol options <vol_Name> no_i2p off
vol options <vol_Name> nvfail off
vol options <vol_Name> read_realloc off

Set up of the 7MTT

These procedures use a Windows based 7-mode transition tool v2.1 (login required).

Moving a data off a ONTAP 8.2.1 7-mode system in to an ONTAP 8.3.1 C-mode switched cluster.

Install the tool on a Windows client machine, preferably a Server OS as it complains it’s not supported on Windows 7 Pro. I have set up a Server based VM running Windows 2012 R2.

Make sure you Active Directory account has been given local admin privileges before logging into the web interface.

It’s recommended to only run the tool against one controller or vfiler at a time, and the assessment and the migrations steps be performed separately.

If the service on the transition host is restarted or the host is rebooted then any system login credentials are lost.