Interface Card Critical Resource Analysis(CRA) Whitepaper for HP-UX 11i v3

Dynamic nPartitions Scenarios

I/O Chassis Deletion Scenarios

Scenario 1

System Configuration:

HP-UX system is running on a cell-based system with multiple cell boards, and

one or more of the cell board(s) is designated as floating cell, with I/O chassis connected to one of

the floating cells.

User Scenario:

To replace or perform hardware maintenance online on a floating cell connected to an

I/O chassis, which has active I/O usage.

To migrate a floating cell connected to an I/O chassis, which has active I/O usage, from one

npartition to another partition dynamically using Dynamic nPars feature.

Operation Performed:

# olrad -D –s <global cell slot id>

To perform online replacement or migration of floating cell, it is necessary to deactivate the I/O

chassis connected to that cell.

The above olrad(1M) command with –D –s option can be used for I/O chassis deactivation.

Criticality reported by CRA:

CRA_SYSTEM_CRITICAL or CRA_DATA_CRITICAL or

CRA_WARNING

Explanation and user action required:

An I/O chassis can be deactivated online, but the deactivation of I/O is subject to critical

resource analysis.

If the I/O chassis about to be deactivated has I/O cards connected to any system or data critical

resources, such as boot disk, disk with a file system, SG quorum disk, NIC with IP address configured,

then the above olrad(1M) operation with –D –s option will fail with either the

CRA_SYSTEM_CRITICAL or CRA_DATA_CRITICAL error.

The actual error depends on whether the resource is system or data critical.

If the CRA reports CRA_WARNING, then the above olrad (1M) chassis deactivation operation will

complete and no specific user action is required.

In case of CRA_SYSTEM_CRITICAL or CRA_DATA_CRITICAL error, if this operation must be

performed for system administration purposes, then, before performing the above operation, ensure

that any active system and data critical resources are reachable through redundant connections

through a different I/O chassis. If there are no redundant connections already available, then

configure the required redundant connections.

However, this recommendation may be applicable only if the applications and services are

configured to use path-independent device special file (DSF) for mass-storage resources or the

networking applications and services are configured with stand-by NICs or Auto-port Aggregation

(APA) group consisting of one or more NICs from a different I/O chassis.

However, if application or service downtime is not an issue for the user, then he may choose, at his

own risk, to stop the applications and services that are using I/O resources through the affected I/O

chassis and then de-configure the critical resource usage.

For example, if there are file systems mounted on devices connected through the affected I/O chassis,

then it is necessary to unmount them. If there are any active LAN cards on the affected I/O chassis, it