Troubleshooting system plan creation for the HMC

Use this information to help resolve problems that you might encounter when creating a system plan with the Hardware Management Console (HMC) Version 7.3.3, and later.

Use HMC V7 R3.3.0, or later versions, to create system plans, as these versions provide the best level of functions for capturing the highest quality and quantity of data from the managed system.

The system plan creation process writes any messages, including error messages to the /var/hsc/log/mksysplan.log. Use the information provided in these tables to determine the type of problem you have and potential solutions for resolving the problem.

The following table contains information about resolving various errors that you might encounter when you create a system plan.

Table 1. Problems and solutions for system plan creation
Problem description	Corrective actions
System plan creation fails with an error that the HMC V 7.3.3.0 failed to obtain the license information for the Virtual I/O Server (VIOS). This error can occur when you are creating a system plan for a system that has a recently installed VIOS and the HMC has not yet established a Resource Monitoring and Control (RMC) connection with the VIOS. This RMC connection is necessary so that the HMC can capture information for the VIOS and its client logical partitions in the system plan.	Use the following command to determine if there is a working RMC connection: /opt/hsc/bin/lspartition -dlpar -c `MTM_MS` `MTM_MS` is machine type, model, and serial number (for example: `9117-MMA_107B730`). If the command results show a value of `1` in the Active field for the specific VIOS partition, the RMC connection is working, and you can redeploy the system plan. The HMC can finish the deployment successfully. If the command results do not indicate that you have a working RMC connection for the VIOS, verify that the system meets the prerequisites for using RMC to further troubleshoot this problem.
Creating a system plan on a POWER6® processor-based server with HMC 7.3.3 fails with an inventory gathering error. You are creating a system plan on a POWER6 processor-based server that has a dual controller card or other unrecognized device on the managed system, and you are using the hardware discovery process to gather information for the new system plan. The hardware discovery process uses firmware code to gather hardware information on the POWER6 processor-based server for unassigned system hardware and hardware that is assigned to an inactive partition to be collected and be included in a system plan. Because the hardware discovery process cannot recognize one or more devices, the HMC fails to create the system plan and the error log indicates an inventory gathering error. The error log might also list location codes that are not valid as they reference planned units and not the actual unit names on the managed system. This problem can occur with dual adapter configurations or if the hardware discovery process finds that a device type other than disk, tape, optical (CD/DVD), or SES is attached to a SCSI, SAS, or SATA controller. PCI adapters that are configured to support dual adapter configurations will be listed as: 572B PCI-X266 Ext Dual-x4 3Gb SAS RAID Adapter 571A PCI-XDDR Dual Channel Ultra320 SCSI Adapter	You must create the system plan without using the hardware discovery option. The system plan still obtains a large amount of information by using the inventory gathering process. If you are using the Create a system plan task from the HMC user interface, make sure that the Retrieve inactive and unallocated hardware resources option is not selected. If you are using the mksysplan command, use one of the following parameters to turn off various data gathering functions for system plan creation, depending on your situation. Note that the –-noprobe parameter turns off both the hardware discovery process and the inventory gathering process: Turn off hardware discovery only. Use --nohwdisc parameter. Turn off inventory gathering only. Use --noinvscout parameter. Turn off information gathering for VIOS only. Use --novios parameter. Turn off both processes. Use –-noprobe parameter.
My system plan creation on HMC 7.3.1, 7.3.2, or 7.3.3 fails with errors that the hardware discovery process was unable to complete because the managed system is not in the Operating or Standby state. This is an example of the type of error messages that occur for this type of deployment failure: Starting inactive partition thickv1 in Hardware Discovery mode. This part usually takes 3 minutes to complete ... Could not complete Hardware Discovery for inactive partition thickv1 because of HSCL0625 Cannot activate partition when the managed system state is not in the Operating or Standby state... This type of failure occurs during system plan creation on a POWER6 processor-based server because the attribute of Power off the system after all the logical partitions are powered off is selected on the General tab of the Managed Systems Properties page. When the hardware discovery process is used to gather information from inactive partitions for the new system plan, the process starts an inactive partition to gather information and powers off the partition when the process is completed. As soon as that partition is ended as part of the hardware discovery process, the managed system will immediately start to power off as required by the selection of this attribute. Consequently, creating the system plan fails because you cannot create a system plan when the managed system is not in either the operating or standby state.	To successfully use the hardware discovery process to create a system plan, ensure that the Power off the system after all the logical partitions are powered off attribute for the managed system is not selected. The hardware discovery process starts partitions and powers off partitions to gather information. If the hardware discovery process powers off the only running partition on the system, the managed system powers off and system plan is not created. To verify this system attribute, complete these steps: In the HMC navigation area, select Systems Management > Servers. In the Tasks area, select Properties. The Properties window for the selected managed system appears. On the General tab, verify that the Power off the system after all the logical partitions are powered off attribute is not selected, and click OK.
System plan creation fails or hangs when the HMC performs the inventory gathering process on an IBM® i partition. On medium to large configurations, a known problem is that the Inventory Scout command line interface (CLI), which the mksysplan command uses, can take several hours to complete.	You must create the system plan without using the inventory gathering process. The system plan still obtains a large amount of information by using the hardware discovery process. When using the mksysplan command, use the --noinvscout parameter to turn off the inventory gathering process. However, running the command this way might result in an incomplete inventory of the system, which affects the quality of the data in the resulting system plan. This problem is fixed in HMC V7 R3.4.0.
After I created my system plan, a logical partition that I did not create, with the name IOR Collection LP exists on my system. How did this partition get on my system and can I delete it? During the hardware discovery process, a new virtual logical partition named IOR Collection LP is created temporarily. This partition is typically deleted by the hardware discovery process right before the mksysplan command is completed. If the mksysplan command or Create system plan task in the HMC completed, and the IOR Collection LP continues to exist after waiting several minutes, report the problem to the HMC technical support.	Contact IBM support, and additionally perform the following steps to delete the IOR Collection LP partition: Write down the partition ID from the partition view of the system on the HMC. Open a terminal connection to the HMC, either on the HMC console or remotely. Use this command: `rmsyscfg -r lpar -m <managed system name> --id <partition id>` For more information, use `rmsyscfg --help` for help on this command.