C Configuration Variables

Table C-1 contains a partial list of configuration variables provided by the TruCluster software products.

Table C-1: Cluster Configuration Variables

Variable Description

ASE When this variable is set to on, the available server environment (ASE) availability services are enabled on the member system.

ASE_ID Specifies a value from 0 to 63 that the cluster software uses to uniquely identify the ASE in which a system resides in a Production Server cluster. Each ASE has a unique ASE identifier; all systems in the same ASE share the same ASE identifier. A member of an ASE in an Available Server configuration always has an ASE_ID of zero (0). See the TruCluster Software Products Software Installation manual for a discussion of ASE identifiers.

ASELOGGER When 1, specifies that the ASE logger daemon (aselogger) is started on the system at boot time. See the TruCluster Software Products Software Installation manual for a discussion of the ASE logger daemon.

ASE_PARTIAL_MIRRORING When this variable is set to on, prevents a service that uses Logical Storage Manager (LSM) mirroring from starting if only one plex of the mirrored data is available.

ASEROUTING Enables host-based routes from a server that has multiple network interfaces. This can make for faster connections to a service by avoiding routers and making use of the multiple interfaces.

CLUSTER_NET Specifies the Internet Protocol (IP) name of the MEMORY CHANNEL interconnect for TruCluster Production Server clusters and TruCluster MEMORY CHANNEL Software configurations, or the hostname in TruCluster Available Server configurations. See the TruCluster Software Products Software Installation manual for more information.

CMS_CONF When set to on, indicates that the Cluster Monitor has been configured on the member system, and that submon and tractd are running.

CNX_DISK Specifies the names of one or more tie-breaker disks used to avoid partitioning in a two-member cluster using a MEMORY CHANNEL connection in virtual hub mode. See Section 2.5 and cnxset(8) for more information.

CNX_INTERVAL Specifies the timeout interval (also known as the maximum ping interval allowed), in seconds, for each member system. A member system times out if it is unable to send at least one ping in this interval. An effective ping interval is derived from this timeout value and sent to the ping daemon (cnxpingd) on each node. The /sbin/init.d/clumember script supplies a default value for this interval when starting the cnxpingd daemon.

CNX_VERBOSE When this variable is set, the connection manager monitor daemon (cnxmond) runs in verbose mode, providing detailed diagnostic messages during its operation. The /sbin/init.d/clumember script supplies a default value for this parameter when starting the cnxmond daemon.

CNX_WAVES Specifies the number of ping intervals (as specified by the -p flag) the connection manager's node monitor daemon should wait before removing from the cluster a member that cannot communicate. The /sbin/init.d/clumember script uses a default value for this parameter.

IMC_AUTO_INIT When this variable is set to 1, the MEMORY CHANNEL API library is automatically initialized at boot time. This initialization involves reserving 0.5 MB for the MEMORY CHANNEL memory the application programming interface (API) library requires. The default value of this attribute is 1.

IMC_MAX_ALLOC Determines the maximum aggregate amount of MEMORY CHANNEL address space the MEMORY CHANNEL API library can allocate for its use across the cluster. If the value of this variable differs among cluster members, the largest value specified on any individual member determines the value set for the cluster. The default amount of address space is 10 MB.

IMC_MAX_RECV Determines the maximum amount of physical memory the MEMORY CHANNEL API library can map for reading MEMORY CHANNEL address space. This limit is node-specific and can vary from member to member. The default amount of address space is 10 MB.

NUM_BIOD Indicates the number of asynchronous I/O servers for distributed raw disk (DRD) services (bsc_biod daemons) that are automatically started at boot time on this member system. See bsc_biod(8) for more information.

TCR_INSTALL Indicates a successful installation when equal to MCS, ASE, orTCR. Indicates an unsuccessful installation when equal to BAD.

TCR_PACKAGE Indicates a successful installation when equal to MCA, ASE, or TCR.

Variable	Description
`ASE`	When this variable is set to `on`, the available server environment (ASE) availability services are enabled on the member system.
`ASE_ID`	Specifies a value from 0 to 63 that the cluster software uses to uniquely identify the ASE in which a system resides in a Production Server cluster. Each ASE has a unique ASE identifier; all systems in the same ASE share the same ASE identifier. A member of an ASE in an Available Server configuration always has an ASE_ID of zero (0). See the TruCluster Software Products Software Installation manual for a discussion of ASE identifiers.
`ASELOGGER`	When 1, specifies that the ASE logger daemon (`aselogger`) is started on the system at boot time. See the TruCluster Software Products Software Installation manual for a discussion of the ASE logger daemon.
`ASE_PARTIAL_MIRRORING`	When this variable is set to `on`, prevents a service that uses Logical Storage Manager (LSM) mirroring from starting if only one plex of the mirrored data is available.
`ASEROUTING`	Enables host-based routes from a server that has multiple network interfaces. This can make for faster connections to a service by avoiding routers and making use of the multiple interfaces.
`CLUSTER_NET`	Specifies the Internet Protocol (IP) name of the MEMORY CHANNEL interconnect for TruCluster Production Server clusters and TruCluster MEMORY CHANNEL Software configurations, or the hostname in TruCluster Available Server configurations. See the TruCluster Software Products Software Installation manual for more information.
`CMS_CONF`	When set to `on`, indicates that the Cluster Monitor has been configured on the member system, and that `submon` and `tractd` are running.
`CNX_DISK`	Specifies the names of one or more tie-breaker disks used to avoid partitioning in a two-member cluster using a MEMORY CHANNEL connection in virtual hub mode. See Section 2.5 and `cnxset`(8) for more information.
`CNX_INTERVAL`	Specifies the timeout interval (also known as the maximum ping interval allowed), in seconds, for each member system. A member system times out if it is unable to send at least one ping in this interval. An effective ping interval is derived from this timeout value and sent to the ping daemon (`cnxpingd`) on each node. The `/sbin/init.d/clumember` script supplies a default value for this interval when starting the `cnxpingd` daemon.
`CNX_VERBOSE`	When this variable is set, the connection manager monitor daemon (`cnxmond`) runs in verbose mode, providing detailed diagnostic messages during its operation. The `/sbin/init.d/clumember` script supplies a default value for this parameter when starting the `cnxmond` daemon.
`CNX_WAVES`	Specifies the number of ping intervals (as specified by the `-p` flag) the connection manager's node monitor daemon should wait before removing from the cluster a member that cannot communicate. The `/sbin/init.d/clumember` script uses a default value for this parameter.
`IMC_AUTO_INIT`	When this variable is set to 1, the MEMORY CHANNEL API library is automatically initialized at boot time. This initialization involves reserving 0.5 MB for the MEMORY CHANNEL memory the application programming interface (API) library requires. The default value of this attribute is 1.
`IMC_MAX_ALLOC`	Determines the maximum aggregate amount of MEMORY CHANNEL address space the MEMORY CHANNEL API library can allocate for its use across the cluster. If the value of this variable differs among cluster members, the largest value specified on any individual member determines the value set for the cluster. The default amount of address space is 10 MB.
`IMC_MAX_RECV`	Determines the maximum amount of physical memory the MEMORY CHANNEL API library can map for reading MEMORY CHANNEL address space. This limit is node-specific and can vary from member to member. The default amount of address space is 10 MB.
`NUM_BIOD`	Indicates the number of asynchronous I/O servers for distributed raw disk (DRD) services (`bsc_biod` daemons) that are automatically started at boot time on this member system. See `bsc_biod`(8) for more information.
`TCR_INSTALL`	Indicates a successful installation when equal to `MCS`, `ASE`, or`TCR`. Indicates an unsuccessful installation when equal to `BAD`.
`TCR_PACKAGE`	Indicates a successful installation when equal to `MCA`, `ASE`, or `TCR`.