Prerequisites for installing HACMP

1. Disk Space Requirements
2. 
   The /usr directory must have 82 MB of free space for a full installation.
   The / (root) directory must have 710 KB of free space
   
   
3. Required AIX BOS Components
4. 
   

<>

AIX BOS Component	AIX v6.1	AIX v5.3
bos.adt.lib	6.1.2.0	5.3.0.1
bos.adt.libm	6.1.2.0	5.3.0.1
bos.adt.syscalls	6.1.2.0	5.3.7.0
bos.net.tcp.client	6.1.2.0	5.3.7.3
bos.net.tcp.server	6.1.2.0	5.3.7.3
bos.rte.SRC	6.1.2.0	5.3.7.3
bos.rte.libc	6.1.2.0	5.3.7.5
bos.rte.libcfg	6.1.2.0	5.3.7.0
bos.rte.libcur	6.1.2.0	5.3.7.0
bos.rte.libpthreads	6.1.2.0	5.3.7.2
bos.rte.odm	6.1.2.0	5.3.7.0
bos.lvm.rte	6.1.2.0	5.3.7.2
bos.clvm.enh	6.1.2.0	5.3.7.2

1. Requirements for NFS v4
2. 
   

The cluster.es.nfs fileset that comes with the HACMP installation medium installs the NFSv4 support for HACMP, along with a NFS Configuration Assistant.

To install this fileset, the following BOS NFS components must also be installed on the system

NFS Fileset	AIX v6.1	AIX v5.3
bos.net.nfs.server	6.1.2.0	5.3.7.0
bos.net.nfs.client	6.1.2.0	5.3.7.0

1. Requirements for RSCT
2. 
   

Install the RSCT images before installing HACMP. Ensure that each node has the same version of RSCT.

To determine if the appropriate filesets are installed and their level, issue the following commands:

# /usr/bin/lslpp -l rsct.compat.basic.hacmp # /usr/bin/lslpp -l rsct.compat.clients.hacmp # /usr/bin/lslpp -l rsct.basic.rte

If these filesets are not present, install the appropriate version of RSCT.
Note: RSCT filesets are available in HACMP as well as AIX installation media.

1. Security fileset requirements
2. 
   

If you plan to use message authentication or encryption for HACMP communication between cluster nodes, the necessary filesets must be installed on each node.

These filessets include:
rsct.crypt.des for data encryption with DES message authentication
rsct.crypt.3des for data encryption standard Triple DES message authentication:
rsct.crypt.aes256 for data encryption with Advanced Encryption Standard (AES) message authentication.

You can install these filesets from the AIX Expansion Pack CD-ROM.

I/O considerations

1. I/O pacing
2. 
   

In certain circumstances, when one application on the system is doing heavy input/output, I/O can take several seconds to complete for another application. During heavy I/O activity, an interactive process can be severely affected if its I/O is blocked or if it needs resources held by a blocked process.

Under these conditions, the HACMP software may be unable to send keepalive packets from the affected node. The RSCT software on other cluster nodes interprets the lack of keepalive packets as node failure, and the I/O-bound node is failed by the other nodes. When the I/O finishes, the node resumes sending keepalives. However, its packets are now out of sync with the other nodes, which then kill the I/O-bound node with a RESET packet.

By default, AIX is installed with high- and low-water marks set to zero, which disables I/O pacing.

Although the most efficient settings for high- and low-water marks vary from system to system, an initial high-water mark of 33 and a low-water mark of 24 provides a good starting point. These settings slightly reduce write times and consistently generate correct fallover behavior from the HACMP software.

1. Syncd frequency
2. 
   

The syncd setting determines the frequency with which the I/O disk-write buffers are flushed. Frequent flushing of these buffers reduces the chance of deadman switch time-outs.

The AIX default value for syncd as set in /sbin/rc.boot is 60. Change this value to 10. Note that the I/O pacing parameter setting should be changed first. You do not need to adjust this parameter again unless time-outs frequently occur.

1. Checking user and group IDs
2. 
   

User and group IDs should be the same on all nodes.

1. Configuring network options
2. 
   

HACMP requires that the nonlocsrcroute , ipsrcroutesend , ipsrcrouterecv , and ipsrcrouteforward network options be set to 1; these are set by RSCT's topsvcs startup script.

If these options are set to anything besides 1, HACMP issues a warning that RSCT will change them.

The verification utility ensures that the value of each network option is consistent across the nodes of a cluster for the following options:


tcp_pmtu_discover
udp_pmtu_discover
ipignoreredirects
routerevalidate.

1. Changing routerevalidate network option
2. 
   

Changing hardware and IP addresses within HACMP changes and deletes routes.

Because AIX caches routes, setting the routerevalidate network option is required as follows:

routerevalidate=1

This setting ensures the maintenance of communication between cluster nodes. To change the default value, add the following line to the end of the /etc/rc.net file:

no -o routerevalidate=1

1. Updating the /etc/hosts file and nameserver configuration
2. 
   

Make sure all nodes can resolve all cluster addresses.

Edit the /etc/hosts file (and the /etc/resolv.conf file, if using the nameserver configuration) on each node in the cluster to make sure the IP addresses of all clustered interfaces are listed.

Also, make sure that the /etc/hosts file on each node has the following entry:

127.0.0.1 loopback localhost

1. Enabling the AIX automounter daemon
2. 
   

For installations that require the AIX automounter daemon on HACMP nodes, a modification is needed to ensure that automounter starts properly (with NFS available and running) on node boot.

This is due to the way HACMP manages the inittab file and run levels upon startup.

To enable the automounter on nodes that have HACMP installed, add the following line as the last line of the file /usr/es/sbin/cluster/etc/harc.net :

chnfs -g on -x 1
startsrc -s nfsd

Note: The chnfs command change is needed only if NFSv4 is being used.