The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, ensure that you understand the potential impact of any command.
The 6248 provides 32 SFP+ universal ports that can operate as 1/10-Gbps fixed Ethernet ports or as 1/2/4/8-Gbps Fibre Channel ports. However, the 6332 provides only 16 SFP+ ports. If you use more than 16 SFP+ ports on the 6248, you must use the breakout ports on the 6332 to meet your requirement for SFP+ ports.
Specifications for the UCS-FI-6332-16UP:
Ports 1 through 16 are SFP+ universal ports that can operate as 1/10-Gbps fixed Ethernet ports or as 4/8/16-Gbps Fibre Channel ports.
Ports 17 through 34 operate as 40-Gbps QSFP+ ports or as eighteen 4 x 10-Gbps SFP+ breakout ports. Alternatively, they can be outfitted with QSA adapters to provide 10-Gbps operation.
Ports 35 through 40 operate as fixed 40-Gbps QSFP+ ports.
Confirm that the current setup consists of the two 6248 FIs in place, configured in a single cluster.
Verify that the cluster high availability (HA) state is ready and that all processes are functioning correctly.
HA should be in the READY state and management services for both FIs should be UP
Note: Verify the migration pre-checks before you begin the migration procedure.
Identify the subordinate 6248 FI in the cluster and evacuate it.
For the subordinate FI:
Remove the connections from the chassis.
Remove the L1-L2 with the other 6248 FI and prepare it for replacement with the new 6332 FI.
Plug in the first 6332 FI with power cables and connect a console cable in order to bring up the first 6332 FI in standalone mode.
Note: You must bring up the first 6332 FI in standalone mode in order to upgrade its firmware to sync it with the current 6248 FI firmware. If you do not bring up the first 6332 FI in standalone mode, the units will not sync their firmware due to the difference in their image types.
Get the GUI up for the standalone FI 6332 and upgrade the infrastructure firmware if it is not already matched.
Connect the 6332 FI to the chassis to ensure that at least one device is connected. This connection provides access to the SEEPROM in order to establish HA between the current FI and the new FI.
Note: This condition is necessary for HA. Without the connected device(s), you will get errors about failed HA even if the L1-L2 connections are done and the firmware is matched.
Configure the ports on the 6332 FI as server and network to establish the connections.
Erase the configuration on the 6332 FI and use the console connection to prepare it to be added to the cluster. Also ensure that L1-L2 are connected.
Use the console to allow the FI to be added as an FI to the current cluster.
Use the console or GUI method for the configuration. Enter the mgmt IP address.
Wait for the FI to come up completely and for HA to be ready.
Verify the pmon state. After the status for both FIs is good, the second FI is ready for replacement.
All processes should be in running state
Proceed with cluster failover to 6332 FI as primary.
Replace the other 6248 FI in a similar manner.
Note: You do not need to upgrade the second 6332 FI in standalone mode. The first 6332 FI can now sync the firmware since the images are the same type and there is not a large difference in the firmware on the two devices.
Make the connections to the second 6332 FI:
Replace the other 6248 FI with the second 6332 FI.
Add the second 6332 FI to the cluster and use the console connection to set it up.
Use the GUI to verify the cluster state.
If server ports on the 6332 FI go into SDP timeout/SFP mismatch error and the error occurs only for one of the two ports connected to each chassis: Leave the connections as they were after multiple unsuccessful attempts to clear the faults. After the 6332 FI is added to the cluster, the errors should clear automatically when the connections are used for discovery.