Recommended Reliable Scalable Cluster Technology (RSCT) package levels for imported AIX images

RSCT is a set of software components that together provide a comprehensive clustering environment for AIX®, Linux®, Solaris, and Windows® operating systems. RSCT is the infrastructure that is used by various IBM products to provide clusters with improved system availability, scalability, and ease of use. For the Power Virtual Server offering, RSCT 3.2.1 is the minimum package level that is required for an imported AIX image (provides IPv6 support). However, the Power Virtual Server development team recommends that you use RSCT 3.2.6.2 for optimal performance.

The RSCT nodeid is not rebuilt if you are deploying an AIX VM from a network installation management (NIM) server without cloud-init, and RSCT is installed.

Resource Management Control (RMC)

The RMC subsystem is the scalable backbone of RSCT that provides a generalized framework for managing resources within a single system or a cluster. Its generalized framework is used by cluster management tools to monitor, query, modify, and control cluster resources. RMC provides a single monitoring and management infrastructure for both RSCT peer domains and management domains.

The AIX VM's RMC status is presented in the Power Virtual Server dashboard as a health status. The RMC health status can be either OK or Warning. Warning statuses happen when the RMC subsystem between your VM and the management system is not connected.

The RMC connection between your VM and the system management service is configured when you create the AIX VM. When you deploy an AIX VM, the IPv6 management interface is injected into the VM. If you remove or overwrite this interface, an RMC connection is not possible. The following procedures can cause the injected IPv6 management interface to be lost after you deploy the AIX VM:

You attach a different boot volume to the VM and boot from it
You use the mksysb restore operation from an on-premises VM
You use smitty to remove the IPv6 interface

Diagnosing and recovering from a missing IPv6 link local address

One of your AIX VM network interface controllers (NICs) must include an IPv6 link local address to connect to the Novalink host. If the desired NIC does not have an associated IPv6 link local address, you must perform a recovery procedure.

Diagnosing a missing IPv6 link local address

Enter the ifconfig -a command on your AIX VM terminal to see whether one of your NICs shows an IPv6 link local address (2001:1234:5723:ABCD:5678:D14E:DBCA:0764/64). The following example is a NIC without an associated IPv6 link local address:

en0: flags=1e084863,480<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTICAST,GROUPRT,64BIT,CHECKSUM_OFFLOAD(ACTIVE),CHAIN>
inet 192.168.2.104 netmask 0xfffffff0 broadcast 192.168.2.111

If one of your NICs does not contain an IPv6 link local address, continue on to the next section.

Using a Power Virtual Server boot image to recover from a missing IPv6 link local address

Enter the lslpp -L rsct.* command to ensure that an operating system (OS) modification did not affect the RSCT file set level. For more informati9on, see Verifying RSCT installation on AIX nodes.

Power Virtual Server supports 3.2.1 ass the minimum release. If you redeployed an AIX image with a package that is older than 3.2.1, you must upgrade RSCT first.
If you still have the Power Virtual Server deployed boot image, complete the following steps:
1. Boot to the original Power Virtual Server boot image.
2. Rerun the ifconfig -a command. The output includes the configured IPv6 link local address.
  
  If you removed the IPv6 link local address from the original boot image configuration, you can read the AIX cloud-init logs to find the IPv6 address.
3. Open the /var/log/cloud-init-output.log file.
4. Use the grep command to search for the IP injection. There is an IPv4 address and an IPv6 link local address.

Using the IPv6 address for the host, refresh the RMC services:

/opt/rsct/bin/rmcctrl -p
/opt/rsct/bin/rmcrefreshMD -s ctrmc

(Optional) If you altered the nodeid (that is unique to RMC), it might have impacted RMC. For example, when you are using PowerHA and trying to copy your nodeid details from an on-premises deployment that is not supported. Begin by rebuilding the node:
```
odmdelete -o CuAt -q name=cluster0 to remove 'cluster0' entry from the CuAt ODM.
```
(Optional) On the AIX VM, enter the cat /etc/ct_node_id command and save the output.
(Optional) To create new nodeid and restart RMC services:
```
/usr/sbin/rsct/install/bin/recfgct
```
(Optional) To build a nodeid, run the /opt/rsct/bin/rmcctrl -p command if not already done in step 3.

If these recovery steps do not restore the RMC status to active and its health to OK, open a case with support.

Recovering from a missing IPv6 link local address when using your own boot image

Complete the following steps to recover from a missing IPv6 link local address:

Before you shut down the AIX VM, grab the IPv6 details.
Gather the IP interface and adapter configuration details by using smitty.
Run the /usr/sbin/rsct/bin/lsnodeid command to grab the nodeid.
Confirm that the nodeid matches the output from cat /etc/ct_node_id and cat /var/ct/cfg/ct_node_id.
Power down the AIX VM and restart it by using your custom boot image.
Re-create the same network configuration as on the IBM boot image by using smitty.
Validate that the RSCT packages are at least 3.2.1 or later by running the lslpp -L rsct.* command.
Run the /usr/sbin/rsct/bin/lsnodeid command. This will likely not match the data from the IBM boot image.

Generate a nodeid to match the original boot ID. Start RMC again and wait 15 minutes.

/usr/sbin/rsct/bin/rmcctrl -p
odmdelete -o CuAt  -q name=cluster0 (optional before the recfgct command)
/usr/sbin/rsct/install/bin/recfgct – I value from /etc/ct_node_id (on the original IBM boot disk collected above)

Run the /usr/sbin/rsct/bin/lsnodeid command.
The output should be the same as the output from the IBM boot image.