Plan for your backup console in the event of hardware or network
failures.
This topic covers backup console information you might want to consider
to quickly recover from the unexpected loss of the console that manages i5/OS®. Many system plans include
a level of redundancy to allow for hardware failures, but some do not consider
the console in those plans. Here are some suggestions for planning for a backup
console to manage i5/OS:
Considerations for a backup console
- The adapter location for Operations Console or a twinaxial console is
fixed, or at least limited for some nonpartitioned IBM System i5™ and eServer™ i5
server models. Based on your server's hardware requirements you might have
limited console choices for i5/OS.
Try to accommodate at least one additional console type, if possible.
- Consider using the takeover and recovery function as part of your i5/OS backup console strategy. However,
the hardware used for the Operations Console or twinaxial console must be
installed prior to the recovery. For more information on the general Operations
Console consideration, see General Operations Console considerations.
- Alternate console refers to a twinaxial console resource
tagged as the alternate console when the console is also a twinaxial console.
An alternate console gives you an added layer of protection because if the
system detects a failure of the primary console during an attended mode IPL,
it automatically tries the alternate console. It is important to remember
that tagging the same resource as both the console and the alternate console
can result in an inability to select a console at all. Consider specifying
the alternate console by tagging an IOA on a different bus in case you development
a bus-related problem.
- IBM System i5 and eServer i5
server models using a Hardware Management Console (HMC) provide you with the ability
to tag a specific IOA as the console device. For more information, see General Operations Console considerations.
- On large multi-partitioned servers or high-availability servers, use Operations
Console (LAN) as the system console for each server or partition. In the event
of a console failure, you can switch to the HMC 5250 console without having
to verify you have additional hardware in place. Most changes of a console
to HMC are performed without the need for an IPL.
Verify your system is ready for a backup console
Recovering
from the loss of a console depends on many factors, some of which include
the model, the hardware resources available, the previous console type, and
the intended new console type. Recovery might consist of repairing the currently
failed console or temporarily replacing it with another console type. Most
changes of a console type can be performed without the need for an IPL, but
there might be circumstances in which an IPL is necessary. Before you use the console service functions (65+21),
verify the following:
- For systems without an HMC,
verify that the console hardware is installed and available.
- For systems with an HMC,
verify that the appropriate console resources are tagged in the partition
profile.
If you plan to use a local console on a network (LAN) as a backup
for another console type, the network adapter must be located in a slot designated
for a console or properly tagged IOA. If not previously configured, you can
use BOOTP to configure the server. For more information
on BOOTP, see Operations Console networking and Creating or verifying a service host name (interface name).
Possible backup console configurations
It is important
to consider as much redundancy as possible for your console needs. If you
consider "what if this fails?" and you have another method to provide a console,
and also make compromises for the hardware requirements necessary to overcome
the various levels where a failure might occur, you are reducing your exposure
to a unrecoverable console failure condition.
Several backup console
configurations can fit your environment. The following table lists possible
solutions.
Table 1. Possible backup console configurations| If... |
Then... |
| If your server is accessed remotely... |
consider an off-site console capability or another
connectivity for the console. A local console on a network (LAN) can be backed
up with additional local console on a network (LAN) PCs. If the network adapter
were to fail, consider a local console that is directly attached as a backup
console. By changing the console type to a local console that is directly
attached with remote access allowed, you can add the ability for a remote
PC to become the console. |
| If your system is configured with an HMC,
but you are running Operations Console... |
you can quickly switch to the HMC 5250
console if Operations Console fails. This can be done without the need to
change hardware. Remember: You must configure each console separately.
|
| If you are using multiple local consoles on a network
(LAN) on a single PC... |
|
| If you are using multiple local consoles on a network
(LAN) on multiple PCs... |
- consider assigning each PC a core set of console responsibilities and
then overlap coverage of backup configurations with each other. For example,
if you have a PC that supports 10 local consoles on a network (LAN) and another
PC with the same number of primary consoles for another 10 logical partitions,
instead of backing up each PC with the other's configuration, you add a third
PC and distribute the 20 consoles so that two PCs back up a portion of each
PC's primary console configurations.
- consider having a dedicated PC to be the backup of a certain number of
consoles, but not connect it until necessary.
|
Note: If more than one local console on a network (LAN) is planned,
create additional service tools device IDs on the server before you start
configuring the Operations Console PC. Each PC connecting to the same target
server or logical partition must have a unique service tools device ID.
For
more information on the ability to switch between console devices, see Switching from one console type to another when a console is currently available.