2 Preinstallation Tasks

This chapter describes the tasks that you perform before loading the latest TruCluster subsets. No matter what type of installation you are performing, read this chapter. If you have not read Chapter 1, please do so before continuing.

To complete the preinstallation tasks, you need access to the following documents:

TruCluster Software Products Release Notes

TruCluster Software Products Hardware Configuration

TruCluster Software Products Administration

A TruCluster product authorization key (PAK) for each license

DIGITAL UNIX Installation Guide

DIGITAL UNIX Release Notes

DIGITAL UNIX System Administration

DIGITAL UNIX Network Administration

Oracle7 Parallel Server Concepts and Administration Guide (if you are performing a rolling upgrade of a Production Server cluster and are running Oracle^® Parallel Server^TM (OPS) software)

Table 2-1 summarizes the preinstallation tasks. It lists the tasks in order, shows the TruCluster products to which each task applies, and provides pointers to necessary information.

Note

For Production Server, a cluster member does not have to be a member of an available server environment (ASE), although there must be at least one ASE in the cluster. If you are installing Production Server on a system that will not be part of an ASE, ignore the tasks related to ASE membership and services.

Table 2-1: Preinstallation Tasks

Task Product See:

Read the TruCluster Software Products Release Notes. PS AS MC TruCluster Software Products Release Notes

Check the hardware and firmware for installation readiness. PS AS MC TruCluster Software Products Hardware Configuration

Obtain any needed IP addresses (MC subnet, ASE services). PS AS MC Section 2.1

Back up the system. PS AS MC DIGITAL UNIX System Administration

If an upgrade, know the value of the system's CLUSTER_NET (cluster interconnect) variable in /etc/rc.config. PS AS Use rcmgr get CLUSTER_NET

If an upgrade, know the value of the system's ASE_ID (ASE ID) variable in /etc/rc.config. PS Use rcmgr get ASE_ID

If an upgrade on a two-system, virtual-hub cluster, know the value of the system's CNX_DISK (tie-breaker disk) variable in /etc/rc.config. PS Use rcmgr get CNX_DISK

If a rolling upgrade and running Oracle^® Parallel Server^TM (OPS) software, stop all OPS database activity on the system. PS Oracle7 Parallel Server Concepts and Administration Guide

If a rolling upgrade, delete this member from the ASE. Run the asemgr utility on a system that is a member of the same ASE as the member you want to delete. You cannot run the asemgr utility on the member system you want to delete. If a simultaneous upgrade, put all services off line. PS AS Section 2.2 and TruCluster Administration

Halt the system and set console variables. PS AS MC Section 2.3

If there are TruCluster subsets on the system, boot /genvmunix to single-user mode and deinstall these subsets. Decide whether or not to save an existing ASE database. PS AS MC Section 2.4

Install the DIGITAL UNIX operating system. PS AS MC Section 2.5 and the DIGITAL UNIX Installation Guide

Configure basic network services (optional for MC). PS AS MC Section 2.6 and the DIGITAL UNIX Network Administration manual

Configure a time service (optional for MC). PS AS MC Section 2.7 and the DIGITAL UNIX Network Administration manual

If performing a simultaneous upgrade and using a manually saved ASE database, restore the ASE database. PS AS Section 2.8

Register one TruCluster product license per system. PS AS MC Section 2.9

Task	Product	See:
Read the TruCluster Software Products Release Notes.	PS	AS	MC	TruCluster Software Products Release Notes
Check the hardware and firmware for installation readiness.	PS	AS	MC	TruCluster Software Products Hardware Configuration
Obtain any needed IP addresses (MC subnet, ASE services).	PS	AS	MC	Section 2.1
Back up the system.	PS	AS	MC	DIGITAL UNIX System Administration
If an upgrade, know the value of the system's `CLUSTER_NET` (cluster interconnect) variable in `/etc/rc.config`.	PS	AS		Use `rcmgr get CLUSTER_NET`
If an upgrade, know the value of the system's `ASE_ID` (ASE ID) variable in `/etc/rc.config`.	PS			Use `rcmgr get ASE_ID`
If an upgrade on a two-system, virtual-hub cluster, know the value of the system's `CNX_DISK` (tie-breaker disk) variable in `/etc/rc.config`.	PS			Use `rcmgr get CNX_DISK`
If a rolling upgrade and running Oracle^® Parallel Server^TM (OPS) software, stop all OPS database activity on the system.	PS			Oracle7 Parallel Server Concepts and Administration Guide
If a rolling upgrade, delete this member from the ASE. Run the `asemgr` utility on a system that is a member of the same ASE as the member you want to delete. You cannot run the `asemgr` utility on the member system you want to delete. If a simultaneous upgrade, put all services off line.	PS	AS		Section 2.2 and TruCluster Administration
Halt the system and set console variables.	PS	AS	MC	Section 2.3
If there are TruCluster subsets on the system, boot `/genvmunix` to single-user mode and deinstall these subsets. Decide whether or not to save an existing ASE database.	PS	AS	MC	Section 2.4
Install the DIGITAL UNIX operating system.	PS	AS	MC	Section 2.5 and the DIGITAL UNIX Installation Guide
Configure basic network services (optional for MC).	PS	AS	MC	Section 2.6 and the DIGITAL UNIX Network Administration manual
Configure a time service (optional for MC).	PS	AS	MC	Section 2.7 and the DIGITAL UNIX Network Administration manual
If performing a simultaneous upgrade and using a manually saved ASE database, restore the ASE database.	PS	AS		Section 2.8
Register one TruCluster product license per system.	PS	AS	MC	Section 2.9

2.1 Obtain IP Names and Addresses (PS, AS, MC)

Depending on which TruCluster product you install, you need to allocate some new Internet Protocol (IP) names and addresses and, for Production Server and MEMORY CHANNEL, a dedicated subnet. The following table shows the address requirements for the TruCluster products.

	Production Server	Available Server	MEMORY CHANNEL
Need dedicated MC subnet with a unique IP address for each system's interface to the subnet	Yes	No	Yes
Need IP addresses for ASE services that require IP addresses to be eligible for failover (these addresses are not on the dedicated MEMORY CHANNEL subnet)	Yes	Yes	Not applicable

The terminology used for network connections by the installation procedures depends on which TruCluster product you install, as follows:

Production Server uses cluster interconnect when referring to a system's MEMORY CHANNEL interface. The IP name and address for this interface are used when establishing a system's membership in a cluster, and for establishing a system's membership in an ASE. Note that Production Server supports redundant MEMORY CHANNEL subnet interfaces (for failover). On any cluster member, these adapters share a single network address; only one IFCONFIG and NETDEV entry represent a system's cluster interconnect in its /etc/rc.config file, regardless of the number of physical MEMORY CHANNEL interfaces present.

Available Server uses member network interface when referring to a system's primary network IP name and address. The IP name and address for this interface are used when establishing a system's membership in an ASE.

MEMORY CHANNEL uses MEMORY CHANNEL adapter when referring to a system's MEMORY CHANNEL interface.

In addition to the terminology distinctions, there is a basic difference between Production Server and Available Server in the network connections between member systems:

Production Server requires a dedicated MEMORY CHANNEL subnet, which is used only for intracluster communication. This MEMORY CHANNEL subnet is the primary network for the cluster; it is a separate and distinct physical network on which only cluster members reside.
You need an IP address (subnet mask 255.255.255.0) and an IP name for the MEMORY CHANNEL cluster interconnect on each member system. See the DIGITAL UNIX Network Administration manual for guidelines on allocating IP addresses. Also see RFC 1918, in which the Internet Assigned Numbers Authority (IANA) reserves the following blocks of IP address space for use by private internets:
```
10.0.0.0      -   10.255.255.255   (10/8 prefix)
172.16.0.0    -   172.31.255.255   (172.16/12 prefix)
192.168.0.0   -   192.168.255.255  (192.168/16 prefix)
```
Note

On the MEMORY CHANNEL subnet, host number 42 is reserved by the Production Server installation procedure as an alias (cluster_cnx) for the connection manager service (cnxmgrd). The installation procedure automatically puts a host entry for cluster_cnx in the /etc/hosts file. You cannot assign host number 42 to another service or member system on the dedicated cluster subnet.

For Available Server, standard network interfaces are used as primary and backup network interfaces. There is no dedicated MEMORY CHANNEL subnet.

For both Production Server and Available Server, the IP addresses used for highly available services must be accessible to clients requesting those services. Therefore, for a Production Server configuration, these addresses cannot be on the dedicated MEMORY CHANNEL subnet. (See the TruCluster Administration manual for information on setting up highly available services.)

Each product's installation procedure updates the /etc/hosts and /etc/rc.config file to reflect the IP names and addresses you supply during installation. For Production Server and Available Server, the installation procedure also assigns the IP name of the primary interface to the CLUSTER_NET variable in /etc/rc.config.

You must manually add entries for the remaining cluster or ASE members to the /etc/hosts file on each system. You can add these entries to /etc/hosts before installing the TruCluster software. You must add these entries before booting the kernel built during the TruCluster installation.

Because the ASE database uses unqualified host names, each entry must include an alias; for example:

123.123.123.123   clu14.abc.def.com  clu14

Note that the system's host name (the one displayed by the hostname program) does not change as a result of installing TruCluster software.

2.2 Prepare ASE Services and Members (PS, AS)

How you deal with available server environment (ASE) services and members depends on whether you are performing a rolling or a simultaneous upgrade. The following sections describe the differences between the two types of upgrade installations.

2.2.1 Rolling Upgrade

To prepare an ASE member for a rolling upgrade, run the asemgr utility on another system that is part of the same ASE and delete the member from the ASE.

Note

The asemgr utility inspects all services in the ASE database that have an automatic service placement (ASP) policy of Favored Member or Restricted to Favored Member, and removes the deleted member from the list of favored members. If this makes any list empty, the following rules apply:

A service with the favored policy effectively becomes one with the balanced policy.

A service with the restricted policy cannot be started until its ASP policy is modified.

2.2.2 Simultaneous Upgrade

For a simultaneous upgrade, run the asemgr utility (on one member system only in each ASE) and perform one of the following tasks:

If you will reuse the existing ASE database, put all services off line.

If you will not reuse the existing ASE database, delete all services.

2.3 Halt System and Set Console Variables (PS, AS, MC)

To halt the system and set the console variables, follow these steps:

Halt the system. For example, to halt the system from multiuser mode with no other users on the system, enter the following command:
```
# shutdown -h now
```

If your system supports the bus_probe_algorithm console variable, set its value to new. This ensures that peripheral component interconnect (PCI) devices are consistently probed on all member systems. To check the setting, enter the following command at the console prompt:
```
>>> show bus_probe_algorithm
bus_probe_algorithm 	new
```
If necessary, enter the following command to set the bus_probe_algorithm variable to new:
```
>>> set bus_probe_algorithm new
```

In order to bring the system to a known state at each reboot, set the boot_reset console variable as follows:
```
>>> set boot_reset on
```

2.4 Boot to Single-User Mode and Deinstall TruCluster Software (PS, AS, MC)

To deinstall existing TruCluster software, follow these steps :

From the console prompt, boot /genvmunix to single-user mode; for example:
```
>>> boot -fl s -fi /genvmunix
```

Enter the bcheckrc command, which makes the root file system writable and mounts local file systems:
```
# /sbin/bcheckrc
```

To make sure that the system's licenses are loaded and active, run the following LMF commands:
```
# lmf reset
# lmf list
```

Use the setld -i command to determine which TruCluster software subsets are installed. The following example shows a Version 1.4A installation, where nnn represents the TruCluster product kit number, for example TCRCONF141:

#  setld -i | grep '^TCR'
 
TCRASEnnn    installed  TruCluster Available Server Software(TruCluster(TM) \
                        Software)
TCRCMSnnn    installed  TruCluster Cluster Monitor (TruCluster(TM) Software)
TCRCOMMONnnn installed  TruCluster Common Components (TruCluster(TM) Software)
TCRCONFnnn   installed  TruCluster Configuration Software (TruCluster(TM) \
                        Software)
TCRDSVCnnn   installed  TruCluster Production Server Software(TruCluster(TM) \
                        Software)
TCRMANnnn    installed  TruCluster Reference Pages (TruCluster(TM) Software)
TCRMCAnnn    installed  TruCluster MEMORY CHANNEL(TM) Software(TruCluster(TM) \
                        Software)

Use the setld -d command to deinstall the subsets.
To ensure that the subsets are deleted in an order that resolves any dependencies between subsets, delete all installed TruCluster subsets with one setld -d command. The following example shows how to delete existing subsets:
```
# setld -d TCRCONFnnn TCRMANnnn TCRCMSnnn TCRDSVCnnn \
TCRMCAnnn TCRASEnnn TCRCOMMONnnn
```
If the deinstallation procedure detects an existing ASE database, /var/ase/config/asecdb, the procedure asks whether you want to save the existing database:
```
Do you want to save the ASE database? [y]
```
For a rolling upgrade, do not save the ASE database. The remaining members of the ASE will repropagate the database when the member is returned to its ASE.
For a simultaneous upgrade, you usually want to save the existing ASE database so you can restart existing services after completing the upgrade.

Caution

If you answer n, the database is deleted. No copy is saved. If you are performing a simultaneous upgrade and delete the database on all systems, you will have to recreate all services after completing the upgrade.

If you save the existing ASE database and are performing a full installation of the DIGITAL UNIX operating system, either back up the database or copy it to another system so that the DIGITAL UNIX Version 4.0D installation does not overwrite the file.

2.5 Install the DIGITAL UNIX Operating System (PS, AS, MC)

Before starting the installation procedures described in the DIGITAL UNIX Installation Guide, read the following list and incorporate these tasks into the installation:

Before installing the operating system, turn on the power to all member systems, the MEMORY CHANNEL hub (if one is included in your configuration), and external disks.

Load all mandatory and any required optional subsets from Table 1-6.

If you are installing the DIGITAL UNIX operating system on a system in an ASE, a disk that is connected to the shared SCSI bus might be listed in the root disk installation menu. Do not install the operating system on a disk that is connected to the shared bus because the data on the disk could be overwritten. DIGITAL recommends that the disks containing the standard system-level file systems (such as root, swap, /usr, and /var) reside on private, nonshared buses.

If you install new hardware (for example, new MEMORY CHANNEL adapters) after you install or update the DIGITAL UNIX operating system, remember to boot /genvmunix and build a customized kernel. Otherwise, the system's kernel configuration file will not contain these hardware options, and the kernel you build during TruCluster installation will not recognize the new hardware. See the DIGITAL UNIX System Administration manual for more information on configuring kernels.

This step applies to systems connected to Asynchronous Transport Mode (ATM) networks. To configure support for ATM LAN Emulation (LANE), select the necessary options from the list displayed by doconfig. In the following partial list of doconfig options, those options required for LANE support are marked with an asterisk (*):
```
     IP Switching over ATM (ATMIFMP)
*    LAN Emulation over ATM (LANE)
     Classical IP over ATM (ATMIP)
*    ATM UNI 3.0/3.1 Signalling for SVCs
```

If the following statements apply to your upgrade path, boot /genvmunix after installing the DIGITAL UNIX operating system and before loading the TruCluster Version 1.5 software:

You are performing an update installation of the DIGITAL UNIX operating system.

You are performing a rolling or simultaneous upgrade of Production Server or Available Server.

Booting /genvmunix before upgrading the TruCluster product removes the possibility of a TruCluster kernel build failure caused by unresolved symbols.

2.6 Configure Basic Network Services (PS, AS, MC)

Using the information in the DIGITAL UNIX Network Administration manual, set up your network and configure basic network services such as the Berkeley Internet Name Domain (BIND) service and the Network Information Service (NIS).

If you configure BIND or NIS:

To configure the members of the cluster as BIND servers, set up one member as a primary server and the rest as secondary servers. To configure the members of the cluster as NIS servers, set up one member as a master server and the rest as slaves. See the information on setting up BIND and NIS servers in the DIGITAL UNIX Network Administration manual.

Because the Cluster Monitor's high-availability feature will not work if cluster members cannot resolve IP addresses, make sure that the hosts entry in /etc/svc.conf has the local service listed before the bind or yp services. For example:
```
hosts=local,bind,yp
```

Make sure that user IDs (UIDs) and group IDs (GIDs) are consistent across all members of the cluster.

2.7 Configure a Time Service (PS, AS, MC)

Set up a distributed time service such as the Network Time Protocol (NTP) daemon (xntpd) on each member system. NTP provides highly accurate synchronization and tracks reliability of time sources. For information on the NTP daemon, see the DIGITAL UNIX Network Administration manual.

Note

If system times are not synchronized, checks that rely on accurate timestamps (for example, in determining which ASE database is the most recent) will fail.
Because the system times of cluster members should not vary by more than a few seconds, DIGITAL recommends that you not use the timed daemon to set the time.

Generally, there is some system in your environment that is considered most informed as far as time is concerned. This system may be getting time from some other system that is considered a reliable time source. If you want the cluster to act as a reliable time source, set up the time service as an ASE service.

2.8 Restore the ASE Database (PS, AS)

If you are performing a simultaneous upgrade, and if you manually moved the ASE database file to another location or system, copy it to its original location: /var/ase/config/asecdb. To set the file's owner to root, group to system, and mode to 644, enter the following commands:

# cd /var/ase/config
# chown root:system asecdb
# chmod 644 asecdb

2.9 Register One TruCluster Software License (PS, AS, MC)

Before you install a TruCluster product, you must use its Product Authorization Key (PAK) to register a product license. (TruCluster licenses are described in Section 1.2.) A PAK is included in your product kit. If you do not have a PAK, make contact with your DIGITAL Customer Services representative.

As shown in the following table, each TruCluster product has a separate PAK:

Product	PAK
TruCluster Production Server Software	TCR-UA
TruCluster Available Server Software	ASE-OA
TruCluster MEMORY CHANNEL Software	MCA-UA

For information on installing a PAK, see the DIGITAL UNIX Software License Management manual, lmf(8), and lmfsetup(8).

Note

You must register the appropriate PAK before installing a TruCluster product; if no PAK is registered, the installation procedure displays the following message:
There are no TruCluster Software licenses installed. In order
to install a TruCluster product you must first install the appropriate
LMF PAK (TCR-UA or MCA-UA or ASE-OA).

2 Preinstallation Tasks

Note

Table 2-1: Preinstallation Tasks

2.1 Obtain IP Names and Addresses (PS, AS, MC)

Note

2.2 Prepare ASE Services and Members (PS, AS)

2.2.1 Rolling Upgrade

Note

2.2.2 Simultaneous Upgrade

2.3 Halt System and Set Console Variables (PS, AS, MC)

2.4 Boot to Single-User Mode and Deinstall TruCluster Software (PS, AS, MC)

Caution

2.5 Install the DIGITAL UNIX Operating System (PS, AS, MC)

2.6 Configure Basic Network Services (PS, AS, MC)

2.7 Configure a Time Service (PS, AS, MC)

Note

2.8 Restore the ASE Database (PS, AS)

2.9 Register One TruCluster Software License (PS, AS, MC)

Note