Standard Tivoli Netcool/OMNIbus automations

A set of standard automations is included with Tivoli Netcool/OMNIbus. These automations are created during database initialization.

The standard automations are stored in the location: $NCHOME/omnibus/etc/automation.sql. You can open the automation.sql file within a text editor and view the syntax of each automation. Comments are included to describe the purpose of the automations. Some of the automations in the automation.sql file are not enabled by default.

You can also use the Netcool/OMNIbus Administrator to browse through these automations by selecting the Automation menu button from the Netcool/OMNIbus Administrator window. You might notice that some trigger groups are disabled by default, whereas the triggers belonging to that trigger group have an enabled state. You must enable such trigger groups in order to run the triggers. One such example is the audit_config trigger group, which provides the ability to raise alerts whenever changes are made to the ObjectServer objects. This trigger group can be used as an audit mechanism in conjunction with the audit log files that are written to the $NCHOME/omnibus/log directory.

The functions performed by some of the standard automations include:
  • Backing up the ObjectServer
  • Adding alerts to the ObjectServer
  • Inserting journal entries
  • Removing redundant entries from various tables

The standard automations are listed in the following table.

Table 1. Standard automations
Trigger or procedure name Description
alert_user_to_old_passwords

Monitors the last recorded time of password change for each user and checks to see if the time is beyond the PasswordMaxAge or, is close to the PasswordMaxAge. If a user has a password that is close to expiring, an event is raised and assigned to that user, to alert them that their password is close to expiring.

Once enabled, the trigger runs every 12 hours. The trigger can be run without enabling the PasswordMaxAge parameter. In this case, users are notified by an event, which is raised when their passwords are approaching, or have exceeded the maximum password age. However, the user is not prevented from logging into the system. For more information, see Configuring the ObjectServer for user authentication.

In an upgraded system, the new trigger is contained in the $OMNIHOME/etc/update81fp16to81fp25.sql update, which must be applied.
audit_config_alter_class Creates an alert indicating that a class has been altered.
audit_config_alter_col_visual Creates an alert indicating that a column visual has been altered.
audit_config_alter_conv Creates an alert indicating that a conversion has been altered.
audit_config_alter_menu Creates an alert indicating that a menu has been altered.
audit_config_alter_object Creates an alert indicating that an object has been altered.
audit_config_alter_prompt Creates an alert indicating that a prompt has been altered.
audit_config_alter_property Creates an alert indicating that a property has been altered.
audit_config_alter_tool Creates an alert indicating that a tool has been altered.
audit_config_create_class Creates an alert indicating that a class has been created.
audit_config_create_col_visual Creates an alert indicating that a column visual has been created.
audit_config_create_conv Creates an alert indicating that a conversion has been created.
audit_config_create_menu Creates an alert indicating that a menu has been created.
audit_config_create_object Creates an alert indicating that an object has been created.
audit_config_create_prompt Creates an alert indicating that a prompt has been created.
audit_config_create_tool Creates an alert indicating that a tool has been created.
audit_config_drop_class Creates an alert indicating that a class has been dropped.
audit_config_drop_col_visual Creates an alert indicating that a column visual has been dropped.
audit_config_drop_conv Creates an alert indicating that a conversion has been dropped.
audit_config_drop_menu Creates an alert indicating that a menu has been dropped.
audit_config_drop_object Creates an alert indicating that an object has been dropped.
audit_config_drop_prompt Creates an alert indicating that a prompt has been dropped.
audit_config_drop_tool Creates an alert indicating that a tool has been dropped.
audit_config_permission_denied Creates an alert indicating that a permission has been denied.
automatic_backup Backs up all ObjectServer memory stores to a sequence of locations dependent on the defined value of a num_backups variable.
automation_disable Disables the automations that should not be running when the ObjectServer is a backup ObjectServer.
automation_enable Enables the automations that should be running when the ObjectServer is a primary ObjectServer.
backup_counterpart_down Enables the automations that should be running when the primary ObjectServer goes down, and the backup ObjectServer is acting as the primary ObjectServer.
backup_counterpart_up Disables the automations that should not be running in the backup ObjectServer when the primary ObjectServer restarts.
backup_failed Specifies an action to perform on a failed backup operation.
backup_startup Disables the automations that should not be running when an ObjectServer designated as a backup, is started.
backup_state_integrity Ensures that only one record is present in the backup state table by cancelling any other inserts.
backup_succeeded Specifies an action to perform on a successful backup operation.
clean_details_table Performs housekeeping cleanup on the alerts.details table. Deletes any entries not found in the alerts.status table.
clean_journal_table Performs housekeeping cleanup on the alerts.journal table. Deletes any entries not found in the alerts.status table.
connection_watch_connect Creates an alert when a new client connects. The process or application name identified by the signal is matched against the alerts.application_types table to identify the appropriate severity and event type for the connect. A gateway connection, for example, is treated as a resolution (clearing a disconnect), whereas an event list connect is a Type 1 event, which will be resolved by a disconnect.
connection_watch_disconnect Creates an alert when a client disconnects. The process or application name identified by the signal is matched against the alerts.application_types table to identify the appropriate severity and event type for the disconnect. A gateway disconnection, for example, is treated as a problem, whereas an event list disconnection is a resolution.
dedup_status_inserts Counts deduplicated status table inserts.
deduplicate_details Deduplicates rows on the alerts.details table.
deduplicate_iduc_stats Deduplicates rows on the iduc_system.iduc_stats table.
deduplication Deduplication processing for the alerts.status table. Maintains the deduplication tally and refreshes alert details.
delete_clears Every 60 seconds, deletes clear alerts that are older than two minutes in the alerts.status table.
details_inserts Counts details table inserts.
disable_inactive_users Runs once a day to disable users who have not logged on to the ObjectServer within a defined period.
disable_user Disables users when they fail to log on after n consecutive failures.
disconnect_iduc_missed Disconnects real-time clients that have not communicated with the ObjectServer for 100 granularity periods.
escalate_off Sets Flash field to 0 (not flashing) and SuppressEscl to 0 (not escalated in this example) when an event that has previously had the Flash field set to 1 is Acknowledged or if the event is Cleared (Severity = 0).
expire Handles the expiration of alerts. Sets the Severity of an alert to 0 if the value of ExpireTime (during which the alert is valid) is exceeded.
flash_not_ack Sets Flashing on (Flash = 1) for events that are 10 minutes old and Critical (Severity = 5), but which have not yet been acknowledged by a user (Acknowledge = 0). It sets SuppressEscl to 1 as a further indication of the escalation status of the event.
generic_clear Clears (Severity = 0) all rows in the alerts.status table indicating a down device (Type = 1), where there is a subsequently inserted row indicating that the device has come back up (Type = 2).
iduc_messages_tblclean Performs housekeeping cleanup on the alerts.iduc_messages table. Runs every 60 seconds and deletes messages older than two minutes.
iduc_stats_insert Inserts a client entry into the iduc_system.iduc_stats table when the iduc_connect signal is raised.
iduc_stats_update Updates the LastIducTime field in the iduc_system.iduc_stats table when the iduc_data_fetch signal is raised.
jinsert Inserts a record into the alerts.journal table. Automations that require journal entries should run this procedure.
journal_inserts Counts journal table inserts.
mail_on_critical Send e-mail about critical alerts that are unacknowledged after 30 minutes.
Note: This tool is UNIX specific unless an equivalent NT mailer is available.
new_row Sets default values for new alerts in the alerts.status table.
new_status_inserts Counts new status table inserts.
pass_deletes Deletes from the destination ObjectServer, rows that do not exist in the source ObjectServer, after resynchronization.
profiler_group_report Writes to the profiler_report file, a row for the sum of the amount of time taken by each distinct application type during the last profiling period.
profiler_report Writes into the profiler_report file, a row for each connected client with the amount of time taken by that client during the last profiling period.
profiler_toggle Reports that the profiler has been toggled.
registry_new_probe A pre-insert trigger that links a probe entry in the registry.probes table with the corresponding entry in the catalog.connections table.
registry_probe_disconnect

Resets the value of the ConnectionID field to 0 in the registry.probes table when a probe or proxy server disconnects from the ObjectServer.
registry_reinsert_probe Updates entries in the registry.probes table in response to reinsert statements that come from a probe. All columns in an entry are updated. The LastUpdate column records the time stamp of the update.
Note: This trigger only accepts reinserts from probes. Other applications must use an UPDATE statement to modify existing entries in the registry.probes table.
registry_update_probe This trigger prevents gateways from overwriting a probe registry entry with outdated information.

This can happen during the fail over or fail back of a probe, when a gateway IDUC cycle occurs after the most recent probe entry has already been written to the ObjectServer.
reset_user Resets the failure count of a user when they log on successfully.
resync_finished Identifies when resynchronization is complete and sets the ActingPrimary property of the backup ObjectServer to FALSE to define it as the backup.
security_watch_security_failure Creates an alert when a client fails to authenticate.
service_insert Service processing for the service.status table.
service_reinsert Service processing for the service.status table.
service_update Service processing for the service.status table.
state_change State change processing for the alerts.status table. Maintains the ObjectServer timestamp of the last insert and update of an alert from any source.
statistics_cleanup Deletes statistics over an hour old.
statistics_gather Collects metrics such as the total number of clients that are connected to the ObjectServer, the number of real-time clients, and the number of new inserts into the alerts.status table, and inserts the metrics into the master.stats table.

This data can then be viewed using Netcool/OMNIbus Administrator or nco_sql, or can be written to file, or processed by other automations.
stats_reset Resets the statistics data.
system_watch_shutdown Creates an alert indicating that the ObjectServer is being shut down.
system_watch_startup Creates an alert indicating that the ObjectServer has started.
trigger_stats_report Writes to the trigger_stats.log file, the amount of time each trigger has used in the last profiling period.
update_service_affecting_events Runs at a specified frequency to enable service-affected events in Network Manager IP Edition to automatically clear when all their related events are cleared. A service-affected event is an alert that warns operators that a critical customer service has been affected by one or more network events.

This automation works only with Tivoli Netcool/OMNIbus V7 .0, or later.

Tip: The automation is required only if Network Manager IP Edition is being used, and is used with the precision.entity_service, precision.service_details, and precision.service_affecting_event tables.
webtop_compatibility Populates the master.profiles table with ObjectServer users for the Web GUI (or Netcool/Webtop) to read. Additionally sets the AllowISQL field for each user who has been granted permission to use the interactive SQL tool in the Web GUI (or Netcool/Webtop).