POWER7 information

Troubleshooting active partition mobility

Learn how to troubleshoot problems that might occur with active partition mobility by using the Hardware Management Console (HMC).

The following table lists possible errors and ways to recover.

Table 1. Known problems and solutions for active partition mobility
Problem Solution
You receive the following error:

HSCL3656 There is an insufficient amount of memory available on the destination managed system for the configuration of the partition. Please perform one or both of the following actions: 1. Remove memory from any shutdown dedicated memory partitions on the destination managed system. 2. Remove memory from any running dedicated memory partitions on the destination managed system.

  1. To make physical memory available for the mobile partition, dynamically remove physical memory from inactive logical partitions that use dedicated memory (subsequently referred to as dedicated memory partitions) on the destination server by running the chhwres command from the HMC command line. For example, chhwres -r mem -m <destination_server> -o r -p <logical_partition> -q <memory>, where:
    • <destination_server> is the name of the server to which you want to move the mobile partition.
    • <logical_partition> is the name of the logical partition from which you want to remove physical memory.
    • <memory> is the amount of physical memory, in MB, that you want to remove from the logical partition.
  2. If you cannot satisfy the memory requirement of the mobile partition by removing physical memory from dedicated memory partitions that are inactive, dynamically remove physical memory from dedicated memory partitions that are active on the destination server by performing one of the following tasks:
You receive the following error:

HSCL03EC There is not enough memory: Obtained : xxxx, Required : xxxx. Check that there is enough memory available to activate the partition. If not, create a new profile or modify the existing profile with the available resources, then activate the partition. If the partition must be activated with these resources, de-activate any running partition(s) using the resource then activate this partition.

  1. To make physical memory available for the mobile partition, dynamically remove physical memory from inactive logical partitions that use dedicated memory (subsequently referred to as dedicated memory partitions) on the destination server by running the chhwres command from the HMC command line. For example, chhwres -r mem -m <destination_server> -o r -p <logical_partition> -q <memory>, where:
    • <destination_server> is the name of the server to which you want to move the mobile partition.
    • <logical_partition> is the name of the logical partition from which you want to remove physical memory.
    • <memory> is the amount of physical memory, in MB, that you want to remove from the logical partition.
  2. If you cannot satisfy the memory requirement of the mobile partition by removing physical memory from dedicated memory partitions that are inactive, dynamically remove physical memory from dedicated memory partitions that are active on the destination server by performing one of the following tasks:
  3. If you cannot satisfy the memory requirement of the mobile partition by dynamically removing physical memory from dedicated memory partitions that are active on the destination server, dynamically remove memory from the mobile partition. For instructions, see Removing dedicated memory dynamically using the HMC.
  4. If you cannot reduce the amount of memory required by the mobile partition to an amount that is equal to or less than the amount of memory that is available on the destination server, shut down logical partitions on the destination server until enough memory is available for the mobile partition to activate on the destination server.
  5. If you cannot satisfy the memory requirement of the mobile partition by shutting down logical partitions on the destination server, move the mobile partition to the destination server by using inactive partition mobility.
 
Notes:
  1. The mobile partition must use dedicated memory. If the mobile partition uses shared memory, skip step 3 and continue to the next step.
  2. After you move the logical partition to the destination server, you might be able to dynamically add one logical memory block (LMB) back to the logical partition. This can happen in one or more of the following situations:
    • The actual available LMBs on the destination server are fractionally high. When determining the available LMBs on the destination server, all fractional LMB sizes are rounded down to the nearest whole number. For example, 5.9 LMBs are rounded down to 5 LMBs.
    • The amount of internal hypervisor storage used on the destination server (to support the logical partition) is a small fraction of 1 LMB. When determining the amount of memory required by the logical partition on the destination server, one LMB is added to the actual number of LMBs required by the logical partition. The added LMB accounts for internal hypervisor storage required to support the logical partition on the destination server.
You receive the following error:

HSCLA319 The migrating partition's virtual Fibre Channel client adapter cannot be hosted by the existing Virtual I/O Server (VIOS) partitions on the destination managed system.

This error indicates that the VIOSs in the target server do not have suitable resources to host the virtual Fibre Channel adapter on the migrating or suspended partition. The following are the most common reasons for this error:
  • The storage area network (SAN) employs port zoning. The target server ports and source server ports are not zoned identically. To host the migrating virtual adapter, the list of Fibre Channel targets in a port on the target server must exactly match the list of Fibre Channel targets in the current mapped port of the migrating virtual adapter on the source server.
  • The two worldwide port names (WWPNs) assigned to the virtual adapter are not zoned identically. The two WWPNs must be interchangeable from both SAN and storage point of view.
  • The target server does not have a port that can meet or exceed the maximum transfer size of the source server port. The maximum transfer size is an attribute of the Fibre Channel port and can be viewed by running the lsattr command on a Fibre Channel device.
  • A switch on the SAN might be configured to use features that extend the Fibre Channel standard in ways that are not compatible with Live Partition Mobility. For example, a port binding feature that tracks WWPN-to-port mappings. This feature can cause problems because Live Partition Mobility validation requires that all ports must be explored through a series of login and logout operations. If the switch tries to track the WWPN-to-port mappings, it might run out of resources and not permit login operations. Disabling this type of feature solves some problems related to failed Fibre Channel login operations.
If the operating system running in the mobile partition does not explicitly support the processor version register of the destination server, and the processor determines that explicit support is required, the processor will not allow the migration to proceed. Perform one of the following actions:
  • Move the logical partition to another system.
  • Update the operating system to a level that supports the target system processor version registers.
You receive an error pertaining to the operating system when you attempt to migrate a logical partition.
  1. Examine the operating system error logs for operating system-related failures.
  2. Examine the HMC log for application-related failures.
You receive an HMC error pertaining to insufficient physical memory on the destination server.
Important: Sufficient physical memory includes the amount of available physical memory on the server and the amount of available contiguous physical memory on the server. If the mobile partition requires more contiguous physical memory, making more physical memory available will not solve the problem.
Perform one of the following actions:
The HMC (or HMCs) and managed system lost their connection while the migration was in progress, or the migration failed. Before running migration recovery ensure that the Resource Monitoring and Control (RMC) connections are established for the migrating partition and the VIOS partitions on the source and destination servers. Complete the following steps from the HMC that manages the source server. If the source server or the source HMC are unavailable, complete the following steps from the HMC that manages the destination server.
  1. In the navigation pane, open Systems Management.
  2. Select Servers.
  3. In the work pane, select the source server. If the source server is unavailable, select the destination server.
  4. In the Tasks menu, select Mobility > Recover. The Migration Recovery window is displayed.
  5. Click Recover.
  6. If you recovered the migration from the HMC that manages the destination server (and a different HMC manages the source server), you might have to manually perform additional recovery tasks on the source server to finish the recovery. For example, even though the migration occurs and the mobile partition runs on the destination server, the mobile partition might appear as an inactive logical partition on the source server. In this situation, remove the mobile partition from the source server to finish the recovery.
Tip: You can also run the migrlpar -o r command to recover a migration.
Note: When you migrate a partition remotely, ensure that you do not connect the source and target servers to the same HMC.
While attempting to change resources dynamically, you receive an error that the RMC daemon is not connected.

This error typically occurs when there is a network connection problem between the logical partitions and the HMC. To resolve this error, check your system network setup.

Live Partition Mobility fails when the client logical partition has multiple virtual Fibre Channel adapters mapped to the same physical Fibre Channel adapter. You cannot migrate or suspend logical partitions that have multiple virtual Fibre Channel adapters mapped to the same physical Fibre Channel adapter.


Send feedback Rate this page

Last updated: Sat, March 17, 2018