cc/td/doc/product/access/sc/r1
hometocprevnextglossaryfeedbacksearchhelp
PDF

Table of Contents

System Management

System Management

TransPath system encompasses much more than the Production platforms that are operating in the field. Many of the subsystems provide support services that are used by customers as part of the overall system. TransPath system provides services to many of these other subsystems. It is important to note that the system architecture is intended to be flexible enough to support both very large and very small customers. Major portions of the system may be distributed over a large geographical area or they may be centralized. TransPath system supports both distributed and centralized architectures equally well.

8.1 Directory Map

This section discusses the directory map of the TransPath system software. All procedures required by the UNIX platform are stored here. For example, syslog.conf or /etc/group is in the "utilities" package. Any re-installation of TransPath system has a reduced impact on the platform reducing the amount of time required to reload packages. Included in this section is an overview of the UNIX file directory tree for the TransPath system software along with a brief description of the purpose for each directory. This section is useful for finding files called out in the operating procedures.

During the installation procedures, the user is asked for a directory under which to install the TransPath system (defaulted to "/opt"). The TransPath directory is created under the base directory and everything in the database is put under it. The following layout assumes TransPath is the base directory of the database. For example, if the user accepts the default installation directory, the base directory would be: "/opt/TransPath/".

Within the TransPath Directory, the following subdirectories are created.

TransPath

/bin

Executable programs provided by Cisco Systems that cannot be customized.

/local

Executable programs provided by Cisco Systems that may be modified by the customer to reflect site-specific requirements. See "Provisioning the Platform." Generally the default values are sufficient.

/data

MDL source files - If MDL source files are provided, they are located here.

/etc

TransPath system configuration files - They include all configuration files required for proper operation of the TransPath system.

/etc/config_lib

TransPath system configuration file library - This is a simple version control system for configuration file changes.

/lib

MDL object files - These are the executables that are created from the MDL source files and dynamically loaded by the Engine at run-time.

/var

Subsystem communication and persistent storage area - This directory contains all files and devices providing communications between the various subsystems of TransPath system. Also contains the files providing persistent storage of data for TransPath system.

/var/log

Subsystem logging area - This directory contains the Command, Alarm, Platform, and the current Dumper logs.

/var/spool

Dumper Spool Area - This directory contains historic Dumper logs.

/var/trace

Signal Path Trace Area - This directory contains all trace logs used for conversion analysis.

8.2 Management Database

This database is a repository that maintains a copy of the configuration data for each managed TransPath system. It stores tailored service packages that have been created by Network Operations personnel. It is also the source of configuration data that is downloaded to each system.

The architecture of the TransPath system application allows the system's configuration to be changed directly in two ways:

The Persistent Database

The Persistent Database is resident on each TransPath system. This database is comprised of the UNIX files located in /opt/TransPath/etc. The only maintenance required is to ensure that regular backups are performed. This database contains all configuration data necessary to support TransPath system. Programs initialize using the data from this database and it allows programs to recover their last known state in the event of a restart.

8.3 XECfgParm.dat

The XECfgParm.dat table defines run-time parameter values that all processes running under the XE can access. This file should be tested during the installation process or when system-wide parameters need to be changed. Please note that the TransPath applications must be stopped and re-started to incorporate the new data.

This table does not contain any relationship-style links to other tables. All parameters currently supported in this table are specified and defined below:

Configuration parameters are defined using the following syntax:

 <facilityName>.<configparameterName> = <parameterValue>

The facilityName identifies a specific processing subsystems of TransPath system. The following are valid facility names:

procM

Process Manager Subsystem

engine

Engine Subsystem

cfgM

Configuration Manager Subsystem

measM

Measurement Manager Subsystem

almM

Alarm Manager Subsystem

ioChanMgr

Cisco IOS Channel Manager

ioChanCtl

Cisco IOS Channel Controllers

MML

Man-Machine Language Subsystem

dmpr

The Log Dumper Subsystem

XE

Execution Environment

foverd

Failover Daemon (for optional configuration only)

*

default facility

If no match is found for a facility-specific parameter, then the default facility is used. For example, the following lines in XECfgParm.dat:

*.logPrio

Notice

engine.logPrio

Debug

The above indicates that all facilities have logging priority set to the Notice level except for the Engine. It has the logging priority set to Debug.

8.4 XECfgParm.dat Configuration Parameter Table

*.platformId

Unique platform identification number (numeric) - Used to allow multiple copies of the platform to run on the same host machine. User can override this value by setting their environment variables.

*.transpathId

Unique TransPath identification number for the ASN - Consists of 1 to 4 hexadecimal digits.

*.homeDirRoot

Root directory path for the platform - Relative path names are derived from this root directory.

*.logPrio

Minimum priority of operational message saved in the logs - Can be changed via the Change Log Level MML command. The following are valid values:

Debug: This priority is used only for detailed debugging messages. 1 - Logging at this priority is not recommended during production, because this causes a large volume of messages to be logged, and degrades system performance.

Info: These messages indicate an operation proceeding as expected. This is the normal logging level for production. This is the default.

Notice: These are status messages generated during the normal course of operation.

Warn: These messages indicate a condition with little or no impact on the operational state of the process, but which may require attention

Error: These messages indicate an unexpected error that is recoverable but degrades performance.

Critical: These messages indicate an error so severe that a process cannot continue operating.

Alert: These messages indicate an error severe enough to impact call processing functions, or failure of the entire system.

Emergency: Not currently used

*.logDest

Logging Destination to which the operational messages are routed. The following are valid values:

file: Local platform file in the "var" directory

stdErr: Standard error I/O stream which defaultsto the local screen. Will not display when the "runAsDaemon" parameter is set.

sysLog: UNIX System Log mechanism.

all: Enable routing to all three of the destinations.

*.logFileFmt

This is the name of log file used when the logDest parameter is set to "file".

If this file does not exist in XECfgParm.dat, then the default log file is "default.log" in the log directory.

*.logFacility

Default message logging facility to write upon - This should always have the following value:

Platform = general message logging

*.tablesFile

Name of the file describing all of the tables of persistent objects that can be loaded.

*.autonomous

Determines whether a process is controlled by the Process Manager. The following are valid values:

true: Not controlled by the Process Manager.

false: Controlled by the Process Manager.

MML.autonomous = true must be set or MML sessions do not operate properly.

Default value used during installation should not be changed. Changes may prevent the Process Manager from controlling a process.

*.runAsDaemon

True/false

If true, the process runs as a daemon. These include the following:

· detaching from controlling terminal.

· becoming a process group leader.

· closing all open files including: stdin, stdout, and stderr.

· opening stdin, stdout, and stderr to /dev/null.

· setting default directory to ../var.

· clearing unmask bits to avoid file protection conflicts.

If false, it will not run as a daemon.

The default value used during installation should not be changed. Changes may prevent the Process Manager from controlling a process.

Engine.LCMMdlFile

Universal Call Model MDL object File

engine.CCMdlFile

Call Context MDL object file

engine.SysCLIval

Provides a default value which indicates whether CLI Validation processing is to be performed on phone calls.

The following are valid values:

true: CLI validation is enabled.

false: CLI validation is not enabled.

engine.SysPropagateChanAvail

The valid value is the following:

true

engine.SysNumTrans

Provides a default value which indicates whether number translation is to be performed on phone calls. The following are valid values:

true: number translation is enabled.

false: number translation is not enabled.

engine.SysToneDetect

Provides a default value which indicates whether Tone Detection (i.e., for in-band analogue signals) is to be performed. Valid values are:

true: tone detection is enabled.

false: tone detection is not enabled.

engine.SysMinOverlap

Positive integer indicating the minimum number of dialed digits allowed on a signalling channel that can carry overlap signalling call initiation messages.

Engine.SysMaxOverlap

Positive integer indicating the maximum number of dialed digits allowed on a signalling channel that can carry overlap signalling call initiation messages.

IoChanMgr.trace

Unit Test tracing of line and channel state machines - It generates a considerable amount of information to the log so set it to "true" only when debugging a particular problem with line or channel service state problems.

IoChanMgr.alarmTimer

Polling period to check alarms in the IO Subsystem. Should be set to zero (the default). This prevents the hardware alarms from being polled.

IoChanMgr.statTimer

Polling period to collect measurements in the IO Subsystem - Should be set around three seconds (30000 milliseconds). Setting to zero (the default) prevents statistics from being polled. This is not necessary for some I/O cards. Please refer to the Product Release Notes.

MeasM.evalInterval

Measurement Manager polling interval in seconds.

procM.minCheckHealthInterval

Minimum interval in seconds, that the procM uses for checking process health. The default value provided during installation should not be changed.

procM.minCheckHealthTimeout

Minimum interval in seconds that the procM allows for timing out on response when health is checked. The default value provided during installation should not be changed.

procM.minKillGracePeriod

Minimum interval in seconds that the procM allows between sending a SIGTERM and sending a SIGKILL when forcing a process to exit. The default value provided during installation should not be changed.

procM.almDwellInterval

Interval in seconds the procM waits before clearing a process alarm. This is used to avoid "bouncing" of alarms. The default value provided during installation should not be changed.

procM.servFmt

Format string used to generate process-specific FIFOs for process management. It has placeholders for platform identification number and process identification number. The default value provided during installation should not need to be changed.

procM.recovDbFile

Determines dbm files used for procM recovery. The default value provided during installation should not need to be changed.

cfgM.recovDbFile

Determines dbm files used for cfgM recovery. The default value provided during installation should not need to be changed.

MML.timeout

Default timeout period for commands in milliseconds - If the command takes longer than this to execute the MML session aborts with an error.

MML.<command>=<time val>

Specific command timeout where <command> is the command name and <time val> is in milliseconds. For example, the following entries in XECfgParm.dat:

MML.timeout = 4000

MML.set-sc-state = 10000

The default values provided during installation are normally optimal. L MML is configured to abort all commands after four seconds except for set-sc-state which aborts after ten seconds.

XE.heartBeatRate

Heartbeat rate in seconds from 5 to 255. Must correspond to the physical setting on the Alarm Relay Unit (ARU).

XE.ARUWriteDevice

Name of the device where the Alarm Relay Unit (ARU) is expecting input from the platform.

foverd.fifo1

Name of FIFO file used by failover daemon for testing purposes. Default value is normally adequate.

foverd.fifo2

Name of FIFO file used by failover daemon for testing purposes. Default value is normally adequate.

foverd.udpPortA

The socket port number used by the failover daemon running on the local system. Should be greater than 1024, and should not conflict with socket ports used by any other applications on the same host.

foverd.udpPortB

The socket port number used by the failover daemon running on the remote system. Should be greater than 1024, and should not conflict with socket ports used by any other applications on the same host.

foverd.udpPortAddrA

The socket IP address used by the failover daemon running on the local system to exchange messages with the remote system.

foverd.udpPortAddrB

The socket IP address used by the failover daemon running on the remote system to exchange messages with the local system.

foverd.asyncPort

The name of the serial port used to exchange messages with the remote system.

foverd.pollTimeout

The frequency at which new I/O messages are processed, in milliseconds. The default value is normally optimal.

foverd.peerCommTimeout

The timeout period after which the remote system is assumed to be unavailable, in milliseconds. Failover will be triggered automatically if there is no communication with the remote system after this timeout period. The default value is normally optimal. Making the timeout value too small will cause false failovers to occur, resulting in possible service interruptions.

foverd.delayTimeout

The amount of delay offset added when the two failover daemons become too closely synchronized. The default value is normally optimal.

foverd.pidFile

The file used to store process information used for recovery purposes. The default value is normally adequate.

foverd.abswitchPort

The serial port used by the local failover daemon to trigger switchover of communication lines when it become the master system.

foverd.procMCmdLine

The command line used by the failover daemon to start the XE Process Manager. This value will only need to be changed if the TransPath software distribution is installed in a non-standard location.

foverd.frepldCmdLine

The command line used by the failoverd daemon to start the file replication daemon. This value will only need to be changed if the TransPath software distribution is installed in a non-standard location.


hometocprevnextglossaryfeedbacksearchhelp
Copyright 1989-1998 © Cisco Systems Inc.