HP Serviceguard Toolkit for Oracle Data Guard manual MAINTENANCEFLAG=yes, Count

Models: Serviceguard Toolkit for Oracle Data Guard

1 140
Download 140 pages 15.58 Kb
Page 134
Image 134

Table 36 User Configuration Variables (continued)

Configuration Variables

Description

MAINTENANCE_FLAG (for example,

This variable will enable or disable maintenance mode

(MAINTENANCE_FLAG=yes)

for the Samba package. By default, this is set to "yes".

 

In order to disable this feature MAINTENANCE_FLAG

 

should be set to "no". When Samba needs to be

 

maintained then a file <package directory>/

 

samba.debug needs to be touched. During this

 

maintenance period Samba process monitoring is

 

paused. Even if Samba is brought down, its package

 

will not be failed over to the standby node. To

 

continue monitoring and come back from the

 

maintenance mode, remove the samba.debug file.

 

It is the user's responsibility to ensure that Samba is

 

properly running after the maintenance phase.

 

NOTE: Setting MAINTENANCE_FLAG to "yes" and

 

touching the samba.debug file in the package

 

directory will put the package in toolkit maintenance

 

mode. Serviceguard A.11.19 release has a new

 

feature which allows individual components of the

 

package to be maintained while the package is still

 

up. This feature is called Package Maintenance mode

 

and is available only for modular packages. For more

 

information using Package Maintenance mode, see

 

white paper Modular package support in Serviceguard

 

for Linux and ECM Toolkits available at http://

 

www.hp.com/go/hpux-serviceguard-docs—>HP

 

Serviceguard Enterprise Cluster Master Toolkit.

MONITOR_INTERVAL (for example, MONITOR_INTERVAL=5)

This variable holds a time interval (in seconds) for

 

monitoring the CIFS Server instance. The monitor

 

process checks to ensure the smbd and nmbd daemons

 

are running every MONITOR_INTERVAL seconds. If

 

not defined, its default value will be 5 seconds.

RETRY_TIMES (for example, RETRY_TIMES=0)

This variable holds the number of times to attempt to

 

check the CIFS Server daemons before giving up and

 

exiting to fail state. If not defined, its default value will

 

be 0 times. It is preferred to keep this as 0.

 

SERVICE_RESTART * RETRY_TIMES = total restart

 

count.

The following information provides the steps for configuring the toolkit and running the package. This includes configuring the SAMBA toolkit user configuration file.

NOTE: Before working on the toolkit configuration, the package directory (example, /etc/ cmcluster/smb1) must be created and all toolkit scripts copied to the package directory.

1.Edit the SAMBA Toolkit user configuration file.

In the package directory, edit the user configuration file (hasmb.conf) as indicated by the comments in that file.

For example:

NETBIOS_NAME=smb1

CONF_FILE=/etc/opt/samba/smb.conf.${NETBIOS_NAME}

LOG_DIRECTORY=/var/opt/samba/${NETBIOS_NAME}/logs

SMBD_PID_FILE=/var/opt/samba/${NETBIOS_NAME}/locks/smbd.pid

NMBD_PID_FILE=/var/opt/samba/${NETBIOS_NAME}/locks/nmbd.pid

2.Copy the entire package directory at the same path on all nodes configured for the package. For this example, each package node must have the following files in the package directory:

134 Using SAMBA Toolkit in a Serviceguard Cluster

Page 134
Image 134
HP Serviceguard Toolkit for Oracle Data Guard MAINTENANCEFLAG=yes, Should be set to no. When Samba needs to be, Count