The PingDirectory server can generate administrative alerts to notify administrators of errors, warnings, and significant events that occur in the server. Each alert has a name, severity, and message.
The docs/admin-alerts-list.csv and docs/admin-alerts-list.html files provide a listing of all alert types that are defined in the PingDirectory server. Some of these alert types include:
- Whenever the server begins or completes the startup process
- Whenever the server begins the shutdown process
- Whenever a change is made to the server configuration
- If the server detects that a configuration change was made with the server offline
- Whenever a change is made to the server’s access control policy
- If a backend is disabled with the server online
- If an error occurs while attempting to enable a connection handler
- If an error occurs while attempting to enable a backend
- If an error occurs while trying to create or restore a backup
- If an error occurs while trying to retrieve or decode an entry from a backend database
- If the server detects that a backend was not cleanly shut down and it can
take additional time to bring it back up while it performs recovery processing
If a backend index needs to be built or is in the process of being built
- If replication has become backlogged, is missing changes, fails to replay a change, encounters a conflict that cannot be automatically resolved, or encounters some other problem
- If the server work queue has become backlogged or full
- If an attempt is made to assign a user an invalid privilege
- If available disk space becomes too low
- If debug logging is enabled
- Whenever the server executes a command on the underlying system
- Whenever the server enters or leaves lockdown mode
- Whenever an alarm is raised or cleared
- If the server detects an invalid or non-recommended Java Virtual Machine (JVM) configuration
- If the PingDirectoryProxy server changes the assessed health check state for any of its backend servers
- If the PingDirectoryProxy server encounters a problem while performing entry rebalancing processing
- If the PingDataSync server detects a backlog or encounters a problem during processing
Whenever an administrative alert is raised within the server, the server provides it to all alert handlers that are defined and enabled in the configuration. Alert handlers that are available in the PingDirectory server include:
- Error Log — Write a message about the alert to server error log.
- Exec — Execute a specified command on the underlying system.
- JMX — Emit a Java Management Extensions (JMX) notification with information about the alert.
- Output — Write a message about the alert to standard output or standard error.
- SMTP — Send an email message with information about the alert to a specified set of recipients.
- SNMP — Emit an SNMP trap with information about the alert.
- Twilio — Send an SMS text message with information about the alert using the Twilio service.
The UnboundID Server SDK can also be used to create custom alert handler implementations.
Information about recent alerts generated by the server is also available in
cn=alerts backend. It is also included in the output of the