Steps to change the IP address on a Prime Infrastructure (PI) server

[ramkris2]

Here are the steps:

If the Prime Infrastructure servers are paired for High Availability (HA), remove HA before performing

these steps. Once the IP address is changed, the servers can be paired again.

1. Login to the console of the Prime Infrastructure server as an 'admin' user.

2. Stop the Service by using 'ncs stop' command.

3. Identify the interface name for which the IP address should be changed using the 'show interface' command

For example:

prime-eagle/admin# show interface 
GigabitEthernet 1
        Link encap:Ethernet  HWaddr 00:50:56:A1:16:3C  
          inet addr:10.10.20.30  Bcast:10.10.20.1 Mask:255.255.255.0
          inet6 addr: 2003:a04::8d3c:2304:51e0:5b1b/64 Scope:Global
          inet6 addr: 2003:a04::250:56ff:fea1:163c/64 Scope:Global
          inet6 addr: fe80::250:56ff:fea1:163c/64 Scope:Link
          inet6 addr: 2003:a04::1089:f316:a0c1:d710/64 Scope:Global
          inet6 addr: 2003:a04::2447:eb8c:c538:2d3/64 Scope:Global
          inet6 addr: 2003:a04::d95f:d46f:e149:3545/64 Scope:Global
          inet6 addr: 2003:a04::c00d:dd7f:84e6:31ce/64 Scope:Global
          inet6 addr: 2003:a04::d17e:c3bd:100f:78ed/64 Scope:Global
          inet6 addr: 2003:a04::8852:f9c0:e4f8:6acf/64 Scope:Global
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:59059145 errors:0 dropped:0 overruns:0 frame:0
          TX packets:32775030 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:27751205364 (25.8 GiB)  TX bytes:15591640503 (14.5 GiB)

In this example, 'GigabitEthernet 1' is the name of the interface.

4. Enter the interface configuration mode to configure the new IP address.

conf t
interface <interface name> 
ip address <new-IP> <subnet-mask>
exit

 After the change, the config is saved to the start-up configuration.

5. Change the Gateway (if needed)

conf t
ip default-gateway <new-gateway-IP-address>
exit
write memory

6. Finally re-start the services using the commands below. It could take anywhere between 15 - 30 minutes for the services to start completely depending on the version of code & allocated resources.

ncs start

Note: Changing the IP address directly from the Linux level  as a 'root' user can cause issues & also won't update the IP address in the Oracle config/listener files.

[Shrikant Gaikwad]

Hello Ramkris2, 

1. do we need to stop the service ncs stop , before changing the ip or we can change it directly?

2.what will happen if we have the two appliances in the HA, do we need to break HA and change the IP address of both appliance and again re-build HA.

[ramkris2]

Hi Shrikant:

Yes, to both your questions.

Ram

[FrostPuppy]

FWIW, changing the IP address of Prime appears to break Controller Configuration Backups. The error messages refer to the TFTP server timing out, which just isn't possible. Digging in the logs shows it's still trying to use the old prime IP address. Running Prime 3.1.5 Update 01 and Device Pack 9

[quark]

The command for th gateway for my part was:

ip default-gateway <ip>

[ramkris2]

That's right, it seems to have changed with newer code. I have updated the gateway command

[David Ritter]

for my part I was not given an option to change the Gateway.. once I changed the IP and said yes.. it was gone.

the last I have was tcl345/admin(config-GigabitEthernet)# ip address 172.25.162.xxx 255.255.255.0

Changing the hostname or IP may result in undesired side effects,
such as installed application(s) being restarted.
Are you sure you want to proceed? [y/n] y

however restart the vm and the app is back on the old ip. to try again.

[Shrikant Gaikwad]

Thanks Everyone

[patoberli]

Just discovered a little bug in PI 3.9.1.

When changing the IP, but staying in the same subnet (for example 192.168.10.20 -> 192.168.10.21), it will remove the default gateway from the internal configuration. It's still shown in "show running", but the server is unreachable. 

To solve it, remove the default gateway (or change it to a different ip address) and add it back. Then the server is reachable again.

[dmuha]

I will note that I just performed the steps here to change the IP of our 3.9 Prime instance to a new IP in the same subnet as the previous IP and I needed to reboot the server to get it to start to ping again.

Stopped service 

Changed interface IP address 

saved

Started Service 

Tried to Ping and access server with no luck

Rebooted 

Ping success to new IP address and was able to access again in HTTPS. 

Figured I'd post incase someone else gets stuck and thinks it didn't work. 

[David Ritter]

I just tried to update the IP of my 3.10.1 instance and after changing the int..

tcl401/admin(config-GigabitEthernet)# ip address 172.25.162.100 255.255.255.0Changing the hostname or IP may result in undesired side effects,such as installed application(s) being restarted.Are you sure you want to proceed? [y/n] y

The system never gave me the opportunity to write mem.  but the VM reported the change.  maybe the GW got flushed.

in any case a restart of the vm restored the package to original IP.  

I will try again and say .'proceed? {y/n} n

doing so aborts the action.    I was doing it from the normal CLI  the 'link' dropped as soon as the IP became 'y'  thus the startup config remained origina.

trying again from the VM door.

[David Ritter]

That worked.  the VM console.  since it was not linked to the IP.