The Database Agent sends the database events to the Controller when the agent detects issues with database connection, failover, or database replication. You can view the details of the events on the Databases > Events page. This page displays the event summary, triggered time, and database type.

You can view the following information in the detailed view:

  • Summary: Displays the details of the event
  • Details: Displays the properties of the event.
  • Actions Executed: Displays the executed actions if you have created a policy.

If you do not want the Database Agent to send database events to the Controller, then set the dbagent.event.trigger.enabled property to false at the agent level. This property is enabled by default.

The following table lists the custom database events with the supported database:

EventDatabase

FAILOVER

  • SQL Server
  • PostgreSQL
  • Oracle
  • MySQL

DB_CONNECTION_DOWN

All database types supported by Database Agent. See Database Visibility Supported Environments.
DB_CONNECTION_UPAll database types supported by Database Agent. See Database Visibility Supported Environments.
REPLICATION_FAILUREPostgreSQL
BLOCKING_SESSIONSQL Server

You can configure alert policies with the custom database events. See Configure Policies.

Failover

Description: This event is sent to the Controller when the primary node fails.

Database TypeSummary
Oracle

Failover occurred in <database-type>, <collector-name> - <database-collector-name> RAC cluster

  • PostgreSQL
  • SQL Server
  • MySQL

Failover occurred for <database-type>, database instance <database-instance-name>

Severity: Error

Custom Event Type: FAILOVER

While configuring alert policies, you can use the Custom Events panel to add a custom event filter.

  1. Click + and specify the custom event type.

  2. Filter the events by using a combination of the following properties:

    PropertyDescription

    availabilityGroupName

    This applies only to SQL server database.

    		Availability group name of the cluster
    failedNodeName of the failed node
    portPort of the node or failed node
    dbCollectorNameName of the database collector
    databaseTypeDatabase type
    primaryNodesAfterFailoverCurrent primary nodes after failover
    timeStampTime at which the event is triggered

Database Connection Down

Description: This event is sent to the Controller when the collector cannot establish a connection with the database or if an existing connection fails. For a cluster, each node failure is sent as a separate event. This event is reported at an exponential time gap: 2, 4, 8, 16, 32 minutes, and so on for 24 hours. Then, the event is reported every 24 hours.

NodeSummary
Single

Connection failure in <database-type>, <database-collector-name, collector-name> 

Cluster

Connection failure in <database-instance-name: node-name>

Severity: Error

Custom Event Type: DB_CONNECTION_DOWN

While configuring alert policies, you can use the Custom Events panel to add a custom event filter.

  1. Click + and specify the custom event type.

  2. Filter the events by using a combination of the following properties:

    PropertyDescription
    dbCollectorNameName of the database collector
    nodeNameName of the node
    databaseTypeDatabase type
    timeStampTime at which the event is triggered
    portPort of the node

Database Connection Up

Description: This event is sent to the Controller when databases get connected after the DB_CONNECTION_DOWN event is triggered.

Severity: Info

Custom Event Type: DB_CONNECTION_UP

While configuring alert policies, you can use the Custom Events panel to add a custom event filter.

  1. Click + and specify the custom event type.

  2. Filter the events by using a combination of the following properties:

    PropertyDescription
    dbCollectorNameName of the database collector
    nodeNameName of the node
    databaseTypeDatabase type
    timeStampTime at which the event is triggered
    portPort of the node

Replication Failure

Description: This event is sent to the Controller when the status property in the pg_stat_replication table changes to stopping or failed.

Summary: Replication failure occurred for database instance - <database-instance-name>, <database-type>

Severity: Error

Custom Event Type: REPLICATION_FAILURE

While configuring alert policies, you can use the Custom Events panel to add a custom event filter.

  1. Click + and specify the custom event type.

  2. Filter the events by using a combination of the following properties:

    PropertyDescription
    confl_bufferpin

    Represents the number of times there was a conflict related to buffer pinning. Buffer pin conflicts occur when multiple transactions attempt to access the same buffer (cache page) concurrently.

    confl_deadlock

    Represents the number of times a deadlock situation was detected in the database. A deadlock occurs when two or more transactions are waiting for each other to release locks, resulting in a circular dependency.

    confl_lock

    Represents the number of times there was a conflict related to locking. Lock conflicts occur when one transaction attempts to acquire a lock that is already held by another transaction.

    confl_snapshot

    Represents the number of times there was a conflict related to concurrent snapshots. A snapshot conflict occurs when a transaction tries to access a row while another transaction is updating or deleting the same row.

    confl_tablespace

    Represents the number of times there was a conflict related to tablespace allocation. A tablespace conflict occurs when multiple transactions try to allocate space in the same tablespace concurrently.

    conninfo

    The connection string used by the WAL receiver process to connect to the primary server.

    databaseTypeDatabase type
    dbCollectorName Name of the database collector
    failedApplicationNodeRepresents the replication slot name used by the WAL receiver process, if applicable. Replication slots are used to track the progress of replication.

    failedNode

    		Name of the node
    last_msg_receipt_timeIndicates the timestamp of the last message received by the standby server from the master server
    last_msg_send_timeIndicates the timestamp of the last message sent from the master server to the standby server
    latest_end_lsn

    Indicates the LSN of the most recent completed WAL record on the standby server. It represents the end of the last applied transaction.

    latest_end_time

    Indicates the timestamp of the most recent completed transaction on the standby server.

    portPort of the node
    reason Reason for the failure
    status

    Indicates the current status of the WAL receiver process. The possible values include:

    • stopped: The WAL receiver is stopped
    • streaming: The WAL receiver is actively streaming WAL data from the master server
    timeStampTime at which the event is triggered

Blocking Session

Description: This event is sent to the Controller when there is a session that is blocked by another session.

Severity: Error

Custom Event Type: BLOCKING_SESSION

While configuring alert policies, you can use the Custom Events panel to add a custom event filter.

  1. Click + and specify the custom event type.

  2. Filter the events by using a combination of the following properties:

    Property

    Description

    blockedObjectThe object that is blocked
    blockedSessionIdID of the session that is blocked
    blockedSqlStatementBlocked query text
    blockedSqlStatementHashHash of the blocked query text
    blockingSessionIdID of the session that is blocking the session
    blockingSqlStatementHashHash of the blocking query text
    databaseTypeDatabase type
    dbCollectorNameName of the database collector
    dbInstanceNameName of the database instance
    duration (hh:mm:ss)

    Duration for which the session is blocked

    eventRuleName

    		Name of the event rule
    eventType

    Name of the reported event

    timeStampTime at which the event is triggered
    violationIDUnique ID of the violation