PowerVM Disk Virtualization (Hybrid) --- vSCSI + NPIV

PowerVM Disk Virtualization

Virtual SCSI
vSCSI is a mechanism that allows the VIOS to present disk volumes to client LPARs across a virtualized SCSI connection. The VIOS owns the physical disk volumes and they can be locally attached or SAN based. The disk volumes are made available to client LPARs using the vSCSI interface and they appear as locally attached SCSI hard disk drives (hdisks) to the client LPAR.

N_Port ID Virtualization
NPIV provides an alternative method for disk virtualization on the VIOS. With NPIV, the physical Fibre-Channel adapter assigned to the VIOS can have 64 downstream virtual World Wide Names (WWN) associated with it. Virtual WWNs can then be assigned to client LPARs

Hybrid vSCSI and NPIV Implementation
VSCSI was the first disk-access virtualization technology provided with PowerVM. It works very well and is in use in many businesses. From a management perspective, VSCSI takes us back to the original UNIX server-implementation model where all disk is presented to the client LPAR as locally attached SCSI disk. The VIOS administrator manages all of the OS and data disk volumes, similar to when servers were configured with local SCSI disk drives. This can become quite a challenging task for servers with large amounts of data, as the responsibility for disk redundancy and backup falls upon their shoulders. One possible solution is to implement a hybrid disk-virtualization model.

With the hybrid model, vSCSI is used for the OS disks and NPIV is used for the data disks. This emulates the traditional environment where there was local SCSI OS disk and Fibre Channel-attached data disk. The client LPAR administrators see local SCSI disk for installation and management of the OS. All of the data disks are presented to the client LPAR through virtual Fibre-Channel connections on the NPIV interface. With regard to disk configuration and management, this allows the client LPAR administrator to focus on managing the OS disk volumes and the SAN administrator to focus on managing the data volumes.

Active Memory Sharing
Active Memory Sharing (AMS) enables the sharing of a pool of physical memory among partitions on a single IBM Power Systems server Power 6 or later, helping to increase memory utilization and drive down system costs.

In order to use the Active Memory Sharing feature of IBM PowerVM, the following are the minimum requirements:

• An IBM Power System server based on the POWER6 processor
• Enterprise PowerVM activation
• Firmware level 340_075
• HMC version 7.3.4 service pack 2 (V7R3.4.0M2) for HMC managed systems
• Virtual I/O Server Version 2.1.0.1-FP21 for both HMC and IVM managed systems
• AIX 6.1 TL 3
• Novell SuSE SLES11
  
  The memory is dynamically allocated amongst the partitions as needed, to optimize the overall physical memory usage in the pool. Instead of assigning a dedicated amount of physical memory to each logical partition, the POWER Hypervisor constantly provides the physical memory from the Shared Memory Pool as needed.
  
  Logical memory:
  Quantity of memory that the operating system manages and can access. Logical memory pages that are in use may be backed up by either physical memory or a pool’s paging device.
  
  For example, four logical partitions with 10 GB of dedicated memory each can be configured to share a memory pool of 40 GB, each with 15 GB of logical memory assigned.
  
  
  Paging:
  A Paging Virtual I/O Server is a partition that provides paging services for a shared memory pool and manages the paging spaces for shared memory partitions associated with a shared memory pool. A Virtual I/O Server enabled as a Paging Virtual I/O Server is designed to serve one shared memory pool.
  
  

This new configuration does not change the global memory requirements, and every logical partition can have the same amount of physical memory it had before. However, memory allocation is highly improved since an unexpected memory demand due to unplanned peak of one logical partition can be satisfied by the shared pool. In deed unused memory pages from shared-memory partitions can be automatically assigned to the more demanding one automatically

The hypervisor has to use a paging device to back up the excess memory that it cannot back up using the physical memory.

A paging device is required for each shared memory partition. The size of the paging device must be equal to or larger than the maximum logical memory defined in the partition profile. The paging devices are owned by a Virtual I/O Server. A paging device can be a logical volume or a whole physical disk. Disks can be local or provided by an external storage subsystem through a SAN.

Reserved storage device pool will be created automatically if AMS will be used, it will need for shared memory paging device.
(Ensure that PVIDs for paging devices for physical volumes set up by the HMC are cleared before use.)

A Virtual Asynchronous Service Interface (VASI) is a virtual device that allows communications between the Virtual I/O Server and the hypervisor. In AMS environment, this device is used for handling hypervisor paging activity.

Active Memory Expansion (AME) -- Power VM

Active Memory Expansion (AME)

Introduction
IBM’s POWER7™ systems with AIX® feature Active Memory™ Expansion, a new technology for expanding a system’s effective memory capacity. Active Memory Expansion employs memory compression technology to transparently compress in-memory data, allowing more data to be placed into memory and thus expanding the memory capacity of POWER7 systems. Utilizing Active Memory Expansion can improve system utilization and increase a system’s throughput.

Active Memory Expansion Overview
Active Memory Expansion relies on compression of in-memory data to increase the amount of data that can be placed into memory and thus expand the effective memory capacity of a POWER7 system. The in-memory data compression is managed by the operating system, and this compression is transparent to applications and users.

Active Memory Expansion is configurable on a per-logical partition (LPAR) basis. Thus, Active Memory Expansion can be selectively enabled for one or more LPAR’s on a system.

When Active Memory Expansion is enabled for a LPAR, the operating system will compress a portion of the LPAR’s memory and leave the remaining portion of memory uncompressed. This results in memory effectively being broken up into two pools – a compressed pool and an uncompressed pool. The operating system will dynamically vary the amount of memory that is compressed based on the workload and the configuration of the LPAR.

The operating system will move data between the compressed and uncompressed memory pools based on the memory access patterns of applications. When an application needs to access data that is compressed, the operating system will automatically decompress the data and move it from the compressed pool to the uncompressed pool, making it available to the application. When the uncompressed pool is full, the operating system will compress data and move it from the uncompressed pool to the compressed pool. This compression and decompression activity is transparent to the application.

Because Active Memory Expansion relies on memory compression, some additional CPU utilization will be consumed when Active Memory Expansion is in-use. The amount of additional CPU utilization needed for Active Memory Expansion will vary based on the workload and the level of memory expansion being used.

System Requirements
Active Memory Expansion is supported across all POWER7 systems. In order to use Active Memory Expansion, the following minimum levels of software are required:

1. HMC: V7R7.1.0.0
2. eFW: 7.1
3. AIX: 6.1 TL4 SP2

VIO -- SMART Admins NIM

 Creating a DVD mksysb form an AIX Lpar

 

On the host you want to restore create an mksysb image with mksysb command :

# mksysb -i /mksysb_images/my_node.mksysb
[..]

You now have to convert this mksysb into CD/DVDs, unfortunatly it’s not possible to create a big iso file and you’ll have to choose CD or DVD format. Use the mkcd command to convert mksysb file into bootable CD/DVDs. In our example we’ll create DVD sized iso files.

# mkcd -L -S -I /mksysb_images/mkcd -m /mksysb_images/my_node.mksysb
Initializing mkcd log: /var/adm/ras/mkcd.log...
Verifying command parameters...
Creating temporary file system: /mkcd/cd_fs...
Populating the CD or DVD file system...
Building chrp boot image...
Copying backup to the CD or DVD file system...
......
Creating Rock Ridge format image: /mksysb_images/mkcd/cd_image_11010246.vol1
Running mkisofs ...
.......
mkrr_fs was successful.

Making the CD or DVD image bootable...

Copying the remainder of the backup to the CD or DVD file system...
Creating Rock Ridge format image: /mksysb_images/mkcd/cd_image_11010246.vol2
Running mkisofs ...
.......
mkrr_fs was successful.

Copying the remainder of the backup to the CD or DVD file system...
Creating Rock Ridge format image: /mksysb_images/mkcd/cd_image_11010246.vol3
Running mkisofs ...
...
mkrr_fs was successful.

• -L : this option is used to create DVD sized iso images.
• -S : this option is used to keep image file, and avoid writing it on a real DVD.
• -I : specify the directory where images will be stored.
• -m : mksysb image file to convert into DVDs.

mkcd command as created three DVD files :

# ls -l
total 53815784
-rw-r--r--    1 root     system   4274950144 Apr 30 13:09 cd_image_22872126.vol1
-rw-r--r--    1 root     system   4293890048 Apr 30 13:12 cd_image_22872126.vol2
-rw-r--r--    1 root     system   4293890048 Apr 30 13:14 cd_image_22872126.vol3

Adding images to VIO Server repository

 

After creating DVDs files from an mksysb file you’ll have to put them on VIO Server repository, transfer it via scp or NFS, and add it into repository :

# mkvopt -name cd_image_22872126.vol1 -file ./cd_image_22872126.vol1
# mkvopt -name cd_image_22872126.vol2 -file ./cd_image_22872126.vol2
# mkvopt -name cd_image_22872126.vol3 -file ./cd_image_22872126.vol3
# lsrep
Size(mb) Free(mb) Parent Pool         Parent Size      Parent Free
   20397     7208 rootvg                   279552           201216

Name                                    File Size Optical         Access
cd_image_22872126.vol1                       4077 None            rw
cd_image_22872126.vol2                       4095 None            rw
cd_image_22872126.vol3                       4095 None            rw

Create Virtual Optic Device on VIO Server

 

On VIO Server create a Virtual Optic Device on Server Virtual SCSI Adapter with mkvdev command

# mkvdev -fbo -vadapter vhost0 -dev lpar_cdrom0
lpar_cdrom0 Available

Loading and Unloading Optic Device

You just have finished, load first DVD on Virtual Optic Device, boot the client on LPAR on Virtual SCSI Adapter and start you restore, then load DVDs one by one :

# loadopt -disk cd_image_22872126.vol1 -vtd lpar_cdrom0 -release
# lsrep
lsrep
Size(mb) Free(mb) Parent Pool         Parent Size      Parent Free
   20397     7208 rootvg                   279552           201216

Name                                    File Size Optical         Access
cd_image_22872126.vol1                       4077 lpar_cdrom0     rw
cd_image_22872126.vol2                       4095 None            rw
cd_image_22872126.vol3                       4095 None            rw

# unloadopt -release -vtd lpar_cdrom0
# loadopt -disk cd_image_22872126.vol2 -vtd lpar_cdrom0 -release
lsrep
Size(mb) Free(mb) Parent Pool         Parent Size      Parent Free
   20397     7208 rootvg                   279552           201216

Name                                    File Size Optical         Access
cd_image_22872126.vol1                       4077 None            rw
cd_image_22872126.vol2                       4095 lpar_cdrom0     rw
cd_image_22872126.vol3                       4095 None            rw

 

Virtual Tape Configuration in VIOS

Step by Step Virtual Tape Drive Configuration

Environment

Virtual Tape supported at VIOS 2.1 with added fixes at VIOS v2.1.0.10-
FP20.1

Resolving the problem

Requirements

VIOS v2.1.0.0 or higher (v2.1.0.10-FP20.1 is the minimum recommended)
Supports SAS-attached tape drives only
New configuration allows physical tape devices to be virtualized

Supported client OS

AIX 5.3 TL 09 or higher
AIX 6.1 TL 03 or higher
LINUX
i5/OS (IBM i)

VIOS partition with NON-target tape (Not a SAS tape drive )

$ lsdev -type tape

name status description
rmt0 Available SCSI 4mm Tape Drive

VIOS Partition with a target SAS tape drive

Step 1:

$ lsdev -type tape

name status description
rmt0 Available SAS 4mm Tape Drive

To list all devices of type "tape" that can be used as backing devices, type

$ lsdev -type tape4vtd

name status description
rmt0 Available SAS 4mm Tape Drive

To export a physical tape drive to a client as a virtual tape drive.

Step 2:

$ mkvdev -vdev rmt0 -vadapter vhost0

vttape0 Available

To list all virtual target tape devices

Step 3:

$ lsdev -type tape -virtual

vttape0 Available Virtual Target Device - Tape

Step 4:

$ lsmap -all -type tape

SVSA Physloc Client Partition ID

-------------- ------------------------- -------------------

vhost0 U8203.E4A.10D4431-V2-C11 0x00000004

VTD vttape0

Status Available

LUN 0x8100000000000000

Backing device rmt0

Physloc U789C.001.DQD2684-P1-C4-T1-L900-L0

Step 5:

Run cfgmgr on client partition to configure the virtual tape drive

Note: If the physical tape drive is currently used as a backing device and you want to use it locally on the VIOS (i.e to create a  backupios), the Virtual Target Device (VTD) must be removed or set in

a Defined state first, or you can also choose to remove the backing device

Example:

To remove VTD, vttape0, type

$ rmvdev -vtd vttape0

To put VTD, vttape0, in a Defined state, type

$ rmdev -dev vttape0 -ucfg

To remove the backing device

$ rmvdev -vdev rmt0

Note: The VIO Server cannot use Atape drive when attempting to Virtualize a SAS tape drive such as the 3580 to AIX client lpar. The Atape driver can be used on the running AIX client lpar.
 

NPIV Setup and Configuration to Virtualize Fibre-channel card on IBM p6

With N_Port ID Virtualization (NPIV), you can configure the managed system so that multiple logical partitions can access independent physical storage through the same physical fibre channel adapter. To access physical storage in a typical storage area network (SAN) that uses fibre channel, the physical storage is mapped to logical units (LUNs) and the LUNs are mapped to the ports of physical fibre channel adapters. Each physical port on each physical fibre channel adapter is identified using one worldwide port name (WWPN). NPIV is a standard technology for fibre channel networks that enables you to connect multiple logical partitions to one physical port of a physical fibre channel adapter. Each logical partition is identified by a unique WWPN, which means that you can connect each logical partition to independent physical storage on a SAN. The figure below illustrates a managed system configured to use NPIV :-   Requirements You can configure virtual fibre channel adapters on client logical partitions that run the following operating systems:   • AIX® version 6.1 Technology Level 2, or later • AIX 5.3 Technology Level 9 • SUSE Linux® Enterprise Server 11, or later   1. Basic requirements. 2. Benefits 3. Disadvantage if any....   1. Basic Requirements will contain: • Advance Power Vitalization License (VET Code) • 8 Gb fibre card. • NPIV supported san switch. • VIO version as 2.1.0 with SP20 and above, it is recommended please updates the VIOS version to latest level. • AIX version 5300-10 and above, Aix 6.1 6100-02-02-0849. • Power 6 hardware • System Firmware version to be 01E#340_075 and above ( where # = L,M,H = Low , Middle , High ) • HMC V7 R3.4.0 and System firmware 340-039   2. Benefits:   • Using NPIV we can assign the LUN's from the storage directly to the VIO client’s. Due to this no mapping will be required to be done from the VIO server. • Usefully when customer is using cluster environment's (HACMP, GPFS). • Unique WWN is generated for each virtual of fiber card. • No additional device driver's filesets needs to be installed on VIO server.   3. Disadvantages:   • NPIV support san switch is mandatory. • 8Gb fiber card can only be virtual • Advance VET license required which leds to additional costs. • Virtual of fiber tape drive is yet not supported.   Configuring Virtual fibre channel for HMC-Managed systems This section describes the steps to create a virtual fibre channel adapter on a p6 LPAR Creating a VFC host adapter on the VIO Server   • Select the VIOS and edit the profile • Go to the “Virtual Adapters” Tab o Select ‘Actions > Create > Fibre Channel Adapter’     • The type of adapter is “Server” since this is being created on the VIOS. • Put a ‘check’ in the box if the adapter is required at the time of partition activation. o You would need this box marked if you are using the adapter for SAN boot. I would recommend always checking it. • Select the client partition you want this Server adapter to be associated with. • Take a note of the assigned Server Adapter ID (13) and the Client Adapter ID (2). You will need to know this information when you create the client adapter. o If the VIO server is already active you will not be able to modify the adapter ID info in the profile once it has been created. Because of this you will need to first check the client for available virtual adapter ID's then when you create the VIO server adapter specify that available ID for the client adapter ID. • Select ‘OK’ to create   You now have a Virtual Fibre Channel (VFC) host adapter created. The next step is to go to you client partition and create a Virtual Fibre Channel client adapter which will be associated with the VFC host adapter. Creating a VFC client adapter on the VIO Client • From the Client Partition that you selected to use when creating the VFC host adapter in the previous steps, go to the “Virtual Adapters” tab then select ‘Actions > Create > Fibre Channel Adapter.. • You will see that the next available Virtual Adapter ID on the the client is “3”. o When we created the VFC host adapter the value was set to “2” for the client adapter ID. So you will have to go back and change the VFC host adapter after the VFC client adapter has been created so the client ID matches. • Put a ‘check’ in the box if the adapter is required at the time of partition activation. o You would need this box marked if you are using the adapter for SAN boot. I would recommend always checking it. • Select the Server partition which has the VFC host adapter that this VFC client adapter will be associated with. • Change the Server Adapter ID to match the Server Adapter ID that you took note of when creating the VFC host adapter. In this case it was “13”. • Select “OK” to create Linking the VFC host and VFC client adapters   • Now back on the VIO Server Partition, Edit the profile and go back to the “Virtual Adapters” tab, select ‘Actions > Edit > Fibre Channel Adapter.. • You will see that the Client Adapter ID is set to “2”. We need to change this to match the adapter ID that was assigned when creating the Client Adapter which was “3”. • Type in the Actual Client Adapter ID and Select “OK” to complete the change. Virtual Fibre Channel Adapter Verification   In the image below is what you should see from both the VIO Server and the Client Partition Profiles.You see the Server has a “Server Fibre Channel Adapter with and ID of “13”, Connecting Partition of“megatron01, and Connecting Adapter of “3” Then from the Client you will see the Client has a “Client Fibre Channel Adapter” with an ID of “3”, a Connecting Partition of “megatronv1”, and a Connecting Adapter of “13”.   • VIO Server: • VIO Client: Creating VFC client adapter on the IVM Client   NOTE: For IVM systems you do not need to create the server adapter and link the server and client adapters yourself. The IVM interface takes care of these two steps for you when you create the client adapter • Open the client properties window. Remember there is no profile for IVM partitions. Click on the "Storage" tab. Find and expand the "Virtual Fibre Channel" section. • If there is not already an entry showing "Automatically generate" under the WWPN column, click on "Add" • Select the new entry, and under the "Physical Port" column select the physical adapter, click ok • After the adapter has been created, you should see the new WWPNs under the "Worldwide Port Names" column • To verify connections and look up client WWPNs, click on the "View Virtual Fibre Channel" option from the main menu • You can see each physical adapter and their connections. Select the adapter and click on "View Partition Connections" to bring up the list of client connections and their WWPNs Mapping the VFC Host Adapter to Physical FC Port In the previous steps we created the virtual fibre channel adapters however they are not yet associated with any physical fibre channel adapter / port. So the next step is to associate the VFC host adapter on the VIO Server with a physical fibre channel adapter port enabling the vfc host adapter to log into the SAN. • Verify the VFC adapters are “Available” on the VIOS $ ioscli lsdev -dev vfchost* name status description vfchost0 Available Virtual FC Server Adapter • Show current VFC mappings to physical (In this example the vfchost has no mappings as indicated by the “NONE”) $ ioscli lsdev -dev vfchost0 -attr map_port NONE N/A True • Display the physical fibre channel adapters (You will decide which adapter / port to use based on how you setup you switch zoning and Storage allocation plan) $ ioscli lsdev -dev fcs* name status description fcs0 Available 8Gb PCI Express Dual Port FC Adapter (df1000f114108a03) fcs1 Available 8Gb PCI Express Dual Port FC Adapter (df1000f114108a03) • Map the virtual fibre channel adapter to the physical fibre channel adapter on the VIOS $ ioscli vfcmap -vadapter “VFCServerAdapter” -fcp “FCPName” Example: ioscli vfcmap –vadapter vfchost0 –fcp fcs0 • Show the VFC mapping to the physical fibre channel adapter ioscli lsdev -dev vfchost0 –attr map_port fcs0 N/A True WWPN, Allocations and Zoning Info When you do your storage allocations you will be allocating the storage directly to the WWPN of the VFC client adapter. If you are attempting to SAN boot, or do Storage allocations prior to the VIO client partition is activated you can display the VFC client adapter WWPN via the HMC with the following steps. • Go to the Client Partition profile and select edit > Select the Virtual Adapters Tab > Highlight the VFC adapter > Select ‘Actions’ > Properties As shown in the picture below, each VFC client adapter will have 2 WWPN’s. When allocating storage you should use the first WWPN in the list. The second WWPN is to be used when performing Partition Migrations (discussed later in this doc). If the VIO client is booted run the following command to display the active WWPN for the VFC client adapter: $ lscfg –vl fcsX <-- where X is the adapter number Network Address.............C05076002A7D0000

SEA Failover Procedure on Dual VIO

 

 How do I setup SEA failover on DUAL VIO servers (VIOS)?  --- is a FAQ

This document provides concepts of SEA (Shared Ethernet Adapter) and the procedure to configure SEA Failover

 

Shared Ethernet Adapter

Shared Ethernet Adapter can be used to connect a physical network to a virtual Ethernet network. It provides the ability for several client partitions to share one physical adapter. SEA can only be configured on the Virtual I/O server (VIOS) and requires the POWER Hypervisor and Advanced POWER Virtualization feature. The SEA, hosted on the Virtual I/O server, acts as a Layer-2 bridge between the internal and external network.

 

Restrictions with Configuring SEA Failover

• It can only be hosted on the VIOS and not on the client partition.
• The VIOS running Integrated Virtualization Manager (IVM) cannot implement SEA Failover because only one single VIOS can be configured on the P5/P6 with IVM.
• SEA Failover was introduced with Fixpack 7 (Virtual I/O server version 1.2), so both Virtual I/O Servers need to be at this minimum level.

Requirements for Configuring SEA Failover

• One SEA on one VIOS acts as the primary (active) adapter and the second SEA on the second VIOS acts as a backup (standby) adapter.
• Each SEA must have at least one virtual Ethernet adapter with the “Access external network” flag (previously known as “trunk” flag) checked. This enables the SEA to provide bridging functionality between the two VIO servers.
• This adapter on both the SEAs has the same PVID, but will have a different priority value.
• A SEA in ha_mode (Failover mode) might have more than one trunk adapters, in which case all should have the same priority value.
• The priority value defines which of the two SEAs will be the primary and which will be the backup. The lower the priority value, the higher the priority, e.g. an adapter with priority 1 will have the highest priority.
• An additional virtual Ethernet adapter , which belongs to a unique VLAN on the system, is used to create the control channel between the SEAs, and must be specified in each SEA when configured in ha_mode.
• The purpose of this control channel is to communicate between the two SEA adapters to determine when a fail over should take place.

NOTE: If the SEA Failover will be using etherchannel as the physical device, configure the switch ports for etherchannel PRIOR to configuring the etherchannel device on the VIO server. Failure to follow this sequence may result in a network storm.

 

PROCEDURE:

1. Create a virtual adapter to be used in the SEA adapter on VIOS1. EX: (ent2).

To configure a virtual Ethernet adapter via Dynamic Logical Partition (DLPAR) for a running logical partition using HMC V7R3, follow these steps:

 

Note: a DLPAR operation requires the partition to be on the network.

1. In the navigation panel, open Systems Management, open Servers, and click on the managed system on which the logical partition is located.
 
2. In the contents panel, select the VIOS on which you want to configure the virtual Ethernet adapter, click on the Tasks button -> choose Dynamic Logical Partitioning -> Virtual Adapters.
 
3. Click Actions -> Create -> Ethernet Adapter.
 
4. Enter the slot number for the virtual Ethernet adapter into Adapter ID.
 
5. Enter the Port Virtual LAN ID (PVID) for the virtual Ethernet adapter into VLAN ID. The PVID allows the virtual Ethernet adapter to communicate with other virtual Ethernet adapters that have the same PVID.
Note: Give the virtual adapter a unique VLAN ID (PVID): "1"
 
6. Select IEEE 802.1 compatible adapter if you want to configure the virtual Ethernet adapter to communicate over multiple virtual LANs. If you leave this option unchecked and you want this partition to connect to multiple virtual networks, then you must create multiple virtual adapters by creating additional virtual LAN IDs.
 
7. Check the box "access external network".
 
8. Give the virtual adapter a low trunk priority. EX: "1"
 
9. Click OK.
 
NOTE !!
After you have finished, access any existing partition profiles for the logical partition and add the same virtual ethernet adapters to those partitions profiles. The dynamic virtual ethernet adapter will be lost if you shut down the logical partition and activate that logical partition using a partition profile that does not have the new virtual ethernet adapter in it.

2. Create another virtual adapter to be used as a Control Channel on VIOS1. EX: (ent3)

a. Give this new virtual adapter another unique VLAN ID (PVID) EX: "99"
b. Do NOT check the box "access external network".
c. Shutdown, Activate VIOS1 or run cfgdev from VIOS command line if created with DLPAR.

3. Create SEA on VIO server 1 with failover attribute:

$ mkvdev -sea ent0 -vadapter ent2 -default ent2 -defaultid 1 -attr ha_mode=auto ctl_chan=ent3
 
Note: The defaultid value of the SEA should be the Port VLAN ID (PVID) of the default trunk adapter, in case there are more than one trunk adapters configured in the SEA.

4. (OPTIONAL) Assign an ip address to SEA on VIOS1:

 

$ mktcpip -hostname vio1 -interface en4 -inetaddr 9.3.5.136 -netmask 255.255.255.0 -gateway 9.3.5.41 -nsrvaddr 9.3.4.2 -nsrvdomain itsc.austin.ibm.com -start

5. Create a virtual adapter to be used in the SEA adapter on VIOS2. EX: (ent2)

a. Give the virtual adapter the same VLAN ID (PVID) as VIOS1. EX: "1" .
b. Check the box "access external network".
c. Give the virtual adapter a higher trunk priority. EX: "2"

6. Create another virtual adapter to be used as a Control Channel on VIOS2. EX: (ent3):

a. Give this new virtual adapter the same unique VLAN ID (PVID) as the control channel on VIOS1. EX: "99"
b. Do NOT check the box "access external network".
c. Shutdown, Activate VIOS2 or run cfgdev from VIOS command line if created with DLPAR.

7. Create SEA on VIOS2 with failover attribute:

$ mkvdev -sea ent0 -vadapter ent2 -default ent2 -defaultid 1 -attr ha_mode=auto ctl_chan=ent3

8. (OPTIONAL) Assign an ip address to SEA on VIOS2:

$ mktcpip -hostname vio2 -interface en4 -inetaddr 9.3.5.137 -netmask 255.255.255.0 -gateway 9.3.5.41 -nsrvaddr 9.3.4.2 -nsrvdomain itsc.austin.ibm.com -start