Upgrade RCF files on Cisco IP switches using CleanUpFiles
You might need to upgrade a reference configuration file (RCF) on a Cisco IP switch. For example, an ONTAP upgrade or a switch firmware upgrade both require a new RCF.
-
Beginning with RcfFileGenerator version 1.4a, there is a new option to change (upgrade, downgrade, or replace) the switch configuration on Cisco IP switches without the need to perform a 'write erase'.
-
Enable console logging before performing this task.
-
The Cisco 9336C-FX2 switch has two different switch storage types that are named differently in the RCF. Use the following table to determine the correct Cisco 9336C-FX2 storage type for your configuration:
If you are connecting the following storage… Choose the Cisco 9336C-FX2 storage type… Sample RCF banner/MOTD -
Directly connected SAS shelves
-
Directly connected NVMe shelves
-
NVMe shelves connected to dedicated storage switches
9336C-FX2 – Direct Storage only
* Switch : NX9336C (direct storage, L2 Networks, direct ISL)-
Directly connected SAS shelves
-
NVMe shelves connected to the MetroCluster IP switches
At least one Ethernet connected NVMe shelf is required 9336C-FX2 – SAS and Ethernet storage
* Switch : NX9336C (SAS and Ethernet storage, L2 Networks, direct ISL) -
You can use this method if your configuration meets the following requirements:
-
The standard RCF configuration is applied.
-
The RcfFileGenerator must be able to create the same RCF that is applied, with the same version and configuration (platforms, VLANs).
-
The RCF that is applied was not provided by NetApp for a special configuration.
-
The RCF was not altered before it was applied.
-
The steps to reset the switch to factory defaults were followed before applying the current RCF.
-
No changes were made to the switch(port) configuration after the RCF was applied.
If you do not meet these requirements, then you cannot use the CleanUpFiles that are created when generating the RCFs. However, you can leverage the function to create generic CleanUpFiles — the cleanup using this method is derived from the output of
show running-configand is best practice.You must update the switches in the following order: Switch_A_1, Switch_B_1, Switch_A_2, Switch_B_2. Or, you can update the switches Switch_A_1 and Switch_B_1 at the same time followed by switches Switch_A_2 and Switch_B_2.
-
Move the cluster LIFs before you begin the upgrade.
-
Determine the current RCF version, and which ports and VLANs are used:
IP_switch_A_1# show banner motdYou need to get this information from all four switches and complete the following information table. * NetApp Reference Configuration File (RCF) * * Switch : NX9336C (SAS storage, L2 Networks, direct ISL) * Filename : NX9336_v1.81_Switch-A1.txt * Date : Generator version: v1.3c_2022-02-24_001, file creation time: 2021-05-11, 18:20:50 * * Platforms : MetroCluster 1 : FAS8300, AFF-A400, FAS8700 * MetroCluster 2 : AFF-A320, FAS9000, AFF-A700, AFF-A800 * Port Usage: * Ports 1- 2: Intra-Cluster Node Ports, Cluster: MetroCluster 1, VLAN 111 * Ports 3- 4: Intra-Cluster Node Ports, Cluster: MetroCluster 2, VLAN 151 * Ports 5- 6: Ports not used * Ports 7- 8: Intra-Cluster ISL Ports, local cluster, VLAN 111, 151 * Ports 9-10: MetroCluster 1, Node Ports, VLAN 119 * Ports 11-12: MetroCluster 2, Node Ports, VLAN 159 * Ports 13-14: Ports not used * Ports 15-20: MetroCluster-IP ISL Ports, VLAN 119, 159, Port Channel 10 * Ports 21-24: MetroCluster-IP ISL Ports, VLAN 119, 159, Port Channel 11, breakout mode 10gx4 * Ports 25-30: Ports not used * Ports 31-36: Ports not used * # IP_switch_A_1#
From this output, you must collect the information shown in the following two tables.
Generic information
MetroCluster
Data
RCF version
1.81
Switch type
NX9336
Network typology
L2 Networks, direct ISL
Storage type
SAS storage
Platforms
1
AFF A400
2
FAS9000
VLAN information Network MetroCluster configuration Switchports Site A Site B VLAN local cluster
Network 1
1
1, 2
111
222
2
3, 4
151
251
Network 2
1
1, 2
111
222
2
3, 4
151
251
VLAN MetroCluster
Network 1
1
9, 10
119
119
2
11, 12
159
159
Network 2
1
9, 10
219
219
2
11, 12
259
259
-
Create the RCFs and CleanUpFiles, or create generic CleanUpFiles for the current configuration.
If your configuration meets the requirements outlined in the prerequisites, select Option 1. If your configuration does not meet the requirements outlined in the prerequisites, select Option 2.
Option 1: Create the RCFs and CleanUpFilesUse this procedure if the configuration meets the requirements.
Steps-
Use the RcfFileGenerator 1.4a (or later) to create the RCFs with the information that you retrieved in Step 1. The new version of the RcfFileGenerator creates an additional set of CleanUpFiles that you can use to revert some configuration and prepare the switch to apply a new RCF configuration.
-
Compare the banner motd with the RCFs that are currently applied. The platform types, switch type, port and VLAN usage must be the same.
You must use the CleanUpFiles from the same version as the RCF and for the exact same configuration. Using any CleanUpFile will not work and might require a full reset of the switch. The ONTAP version the RCF is created for is not relevant. Only the RCF version is important. The RCF (even if it is the same version) might list fewer or more platforms. Make sure that your platform is listed.
Option 2: Create generic CleanUpFilesUse this procedure if the configuration does not meet all the requirements.
Steps-
Retrieve the output of
show running-configfrom each switch. -
Open the RcfFileGenerator tool and click 'Create generic CleanUpFiles' at the bottom of the window
-
Copy the output that you retrieved in Step 1 from 'one' switch into the upper window. You can remove or leave the default output.
-
Click 'Create CUF files'.
-
Copy the output from the lower window into a text file (this file is the CleanUpFile).
-
Repeat Steps c, d, and e for all switches in the configuration.
At the end of this procedure, you should have four text files, one for each switch. You can use these files in the same way as the CleanUpFiles that you can create by using Option 1.
-
-
Create the 'new' RCFs for the new configuration. Create these files in the same way that you created the files in the previous step, except choose the respective ONTAP and RCF version.
After completing this step you should have two sets of RCFs, each set consisting of twelve files.
-
Download the files to the bootflash.
-
Download the CleanUpFiles that you created in Create the RCFs and CleanUpFiles, or create generic CleanUpFiles for the current configuration
This CleanUpFile is for the current RCF that is applied and NOT for the new RCF that you want to upgrade to. Example CleanUpFile for Switch-A1:
Cleanup_NX9336_v1.81_Switch-A1.txt -
Download the 'new' RCFs that you created in Create the 'new' RCFs for the new configuration.
Example RCF for Switch-A1:
NX9336_v1.90_Switch-A1.txt -
Download the CleanUpFiles that you created in Create the 'new' RCFs for the new configuration. This step is optional — you can use the file in future to update the switch configuration. It matches the currently applied configuration.
Example CleanUpFile for Switch-A1:
Cleanup_NX9336_v1.90_Switch-A1.txtYou must use the CleanUpFile for the correct (matching) RCF version. If you use a CleanUpFile for a different RCF version, or a different configuration then the cleanup of the configuration might not work correctly. The following example copies the three files to the bootflash:
IP_switch_A_1# copy sftp://user@50.50.50.50/RcfFiles/NX9336-direct-SAS_v1.81_MetroCluster-IP_L2Direct_A400FAS8700_xxx_xxx_xxx_xxx/Cleanup_NX9336_v1.81_Switch-A1.txt bootflash: IP_switch_A_1# copy sftp://user@50.50.50.50/RcfFiles/NX9336-direct-SAS_v1.90_MetroCluster-IP_L2Direct_A400FAS8700A900FAS9500_xxx_xxx_xxx_xxxNX9336_v1.90//NX9336_v1.90_Switch-A1.txt bootflash: IP_switch_A_1# copy sftp://user@50.50.50.50/RcfFiles/NX9336-direct-SAS_v1.90_MetroCluster-IP_L2Direct_A400FAS8700A900FAS9500_xxx_xxx_xxx_xxxNX9336_v1.90//Cleanup_NX9336_v1.90_Switch-A1.txt bootflash:
You are prompted to specify Virtual Routing and Forwarding (VRF).
-
-
Apply the CleanUpFile or generic CleanUpFile.
Some of the configuration is reverted and switchports go 'offline'.
-
Confirm that there are no pending changes to the startup configuration:
show running-config diffIP_switch_A_1# show running-config diff IP_switch_A_1#
-
-
If you see system output, save the running configuration to the startup configuration:
copy running-config startup-configSystem output indicates that the startup configuration and running configuration are different and pending changes. If you do not save the pending changes, you are unable to roll back using a reload of the switch. -
Apply the CleanUpFile:
IP_switch_A_1# copy bootflash:Cleanup_NX9336_v1.81_Switch-A1.txt running-config IP_switch_A_1#
The script might take a while to return to the switch prompt. No output is expected.
-
-
View the running configuration to verify that the configuration is cleared:
show running-configThe current configuration should show:
-
No class maps and IP access lists are configured
-
No policy maps are configured
-
No service policies are configured
-
No port-profiles are configured
-
All Ethernet interfaces (except mgmt0 which should not show any configuration, and only VLAN 1 should be configured).
If you find that any of the above items are configured, you might not be able to apply a new RCF configuration. However, you can revert to the previous configuration by reloading the switch without saving the running configuration to the startup configuration. The switch will come up with the previous configuration.
-
-
Apply the RCF and verify that the ports are online.
-
Apply the RCF.
IP_switch_A_1# copy bootflash:NX9336_v1.90-X2_Switch-A1.txt running-config
Some warning messages appear while applying the configuration. Error messages are generally not expected. However, if you are logged in using SSH, you might receive the following error: Error: Can't disable/re-enable ssh:Current user is logged in through ssh -
After the configuration is applied, verify that the cluster and MetroCluster ports are coming online with one of the following commands,
show interface brief,show cdp neighbors, orshow lldp neighborsIf you changed the VLAN for the local cluster and you upgraded the first switch at the site, then cluster health monitoring might not report the state as 'healthy' because the VLANs from the old and new configurations do not match. After the second switch is updated, the state should return to healthy. If the configuration is not applied correctly, or you do not want to keep the configuration, you can revert to the previous configuration by reloading the switch without saving the running configuration to startup configuration. The switch will come up with the previous configuration.
-
-
Save the configuration and reload the switch.
IP_switch_A_1# copy running-config startup-config IP_switch_A_1# reload
-
Modify all cluster LIFs to enable auto-revert:
-
Re-enable auto-revert:
network interface modify -vserver Cluster -lif * -auto-revert truecluster_A::*> network interface modify -vserver Cluster -lif * -auto-revert true Warning: Changes to a cluster LIF can affect the health of the Cluster. Are you sure you want to continue? {y|n}: y 4 entries were modified. cluster_A::*> -
Revert all cluster LIFs that are not currently on their home ports to their home ports:
network interface revert -vserver Cluster -lif *Wait a few minutes for the LIFs to revert to their home port:
-
Confirm that all LIFs are on their home ports:
cluster_A::*> network interface show -role cluster Logical Status Network Current Current Is Vserver Interface Admin/Oper Address/Mask Node Port Home ----------- -------------- ---------- ------------------ ------------- ------- ---- Cluster node_A_1_clus1 up/up 192.168.204.88/18 node_A_1 e4a true node_A_1_clus2 up/up 192.168.204.93/18 node_A_1 e4b true node_A_2_clus1 up/up 192.168.204.96/18 node_A_2 e4a true node_A_2_clus2 up/up 192.168.204.97/18 node_A_2 e4b true 4 entries were displayed. -
Verify the state of the MetroCluster nodes:
metrocluster node showcluster_A::*> metrocluster node show DR Configuration DR Group Cluster Node State Mirroring Mode ----- ------- ------------------ -------------- --------- -------------------- 1 cluster_A node_A_1 configured enabled normal node_A_2 configured enabled normal cluster_B node_B_1 configured enabled normal node_B_2 configured enabled normal 4 entries were displayed. cluster_A::*>
-