Serviceguard NFS Toolkit Support for Cluster File System (October 2011)

NODE STATUS STATE

thyme up running

NODE STATUS STATE

basil up running

MULTI_NODE_PACKAGES

PACKAGE STATUS STATE AUTO_RUN SYSTEM

SG-CFS-pkg up running enabled yes

SG-CFS-DG-1 up running enabled no

SG-CFS-MP-1 up running enabled no

SG-CFS-MP-2 up running enabled no

SG-NFS-XP-1 up running enabled no

Configuring a Serviceguard NFS failover package

Configuring a Serviceguard NFS failover package for a CFS environment is similar to

configuring the package for a non-CFS environment. The main difference is that you

must configure one failover package for every server that exports CFS.

1. Create the nfs.conf file for each package with the cmmakepkg command. You must

create one package for each server.

# cd /etc/cmcluster/nfs_modular

# cmmakepkg -m sg/all -m nfs/hanfs /etc/cmcluster/nfs_modular/\

nfs1.conf

# cmmakepkg -m sg/all -m nfs/hanfs /etc/cmcluster/nfs_modular/\

nfs2.conf

2. Edit the nfs.conf files (nfs1.conf, nfs2.conf) as follows:

a) Specify the IP address for the package and the subnet to which the IP address

belongs:

ip_subnet 15.13.112.0

ip_address 15.13.114.243

This IP address is the relocatable IP address for the package. NFS clients that mount

the file systems in the package will use this IP address to identify the server. You must

configure a name for this IP address in the DNS, NIS, or LDAP database, or in the

/etc/hosts file.

b) By default the service_name is enabled and is set to nfs.monitor. To monitor NFS

services, change nfs.monitor to a unique name for each package.

service_name nfs1.monitor

service_cmd "$SGCONF/scripts/nfs/nfs_upcc.mon"

service_restart none

service_fail_fast_enabled no

service_halt_timeout 0

In the nfs2.conf file, set service_name to nfs2.monitor. If you do not want to monitor

NFS services, comment out the service_name, service_cmd, service_restart,

service_fail_fast_enabled and service_halt_timeout variables