Warning

License: Pro - Requires a Pro or Enterprise license.

Overview

The Oracle Enterprise probe monitors Oracle Database instances that run Enterprise Edition with the Diagnostics Pack, collecting performance and availability metrics from the in-memory diagnostic views: v$sysmetric (AWR-equivalent 1-minute rates), v$active_session_history (ASH), the gv$ cluster views for Real Application Clusters (RAC), and v$dataguard_stats for Data Guard standby lag.

One probe instance monitors one Oracle service (SID / PDB); add more instances for additional databases. The monitored instance is also emitted as a db entity so it appears in the topology view, anchored from the moment the probe starts even while the database is unreachable.

Diagnostics Pack licensing

v$sysmetric and v$active_session_history are part of the Oracle Diagnostics Pack, a separately licensed option of Enterprise Edition. For a Standard Edition or unlicensed instance, use the Free Oracle Database probe instead, which reads only the always-free dynamic performance views.

Collected data:

• Instance reachability
• AWR-equivalent per-second rates: DB time, DB CPU, hard/soft parses, logical and physical reads/writes, executions
• Active Session History: active sessions by wait class, and sessions on CPU
• RAC (optional): open instance count and per-instance SQL*Net traffic and global-cache blocks received
• Data Guard: redo apply lag and transport lag on the standby

All metrics are emitted under the senhub.oracle_enterprise.* namespace. Wait-class and per-RAC-instance series carry attributes (wait_class, rac.instance) that keep instances distinct in OTLP/Prometheus and act as filters in the Sensor URLs tab of the console.

Quick Start

Basic Configuration

# probes.d/40-oracle-enterprise.yaml — each file under probes.d/ is a YAML array of probes
- name: oracle-enterprise-prod
  type: oracle_enterprise
  params:
    dsn: "oracle://monitoring:${secret:oracle-prod.password}@oracle.company.com:1521/ORCLPDB1"
    interval: 300
    timeout: 30
    collect_rac: true

The dsn is a standard Oracle connection URL — oracle://user:password@host:port/service. The ${secret:...} reference resolves the password from the OS-native secret store (see Configuration); inline plaintext is auto-sealed on install.

Single-instance database (no RAC)

Disable the gv$ cluster queries on a non-clustered instance to avoid needless work:

# probes.d/40-oracle-enterprise.yaml
- name: oracle-enterprise-single
  type: oracle_enterprise
  params:
    dsn: "oracle://monitoring:${secret:oracle-single.password}@db1.company.com:1521/XEPDB1"
    interval: 300
    collect_rac: false

Configuration Parameters

Parameter	Must set	Default	Description
dsn	Yes	-	Connection string carrying user, password, host, port and service name. A secret: reference it with ${secret:…}, ${env:…} or ${file:…} rather than writing it in the file. Example: oracle://monitor:secret@db01.example.com:1521/ORCL
interval	No	300	Seconds between collections
timeout	No	30	Query timeout per cycle, in seconds
collect_rac	No	true	Query the gv$ cluster views for RAC metrics; false on a single-instance database

Metrics Collected

All metrics carry engine, instance, server.address, server.port and service attributes identifying the database. Families that split by a dimension (wait class, RAC instance) use an OTel attribute rather than separate metric names.

Overview

Metric	Unit	Description
senhub.oracle_enterprise.up	1	1 when the instance answered this cycle, else 0

AWR (v$sysmetric)

Metric	Unit	Description
senhub.oracle_enterprise.awr.db_time	s	Database time consumed per second
senhub.oracle_enterprise.awr.db_cpu	s	Database CPU time consumed per second
senhub.oracle_enterprise.awr.parse.hard	{parse}/s	Hard parses per second
senhub.oracle_enterprise.awr.parse.soft	{parse}/s	Soft parses per second
senhub.oracle_enterprise.awr.logical_reads	{read}/s	Logical reads per second
senhub.oracle_enterprise.awr.physical_reads	{read}/s	Physical reads per second
senhub.oracle_enterprise.awr.physical_writes	{write}/s	Physical writes per second
senhub.oracle_enterprise.awr.executions	{execution}/s	SQL executions per second

ASH (v$active_session_history)

Metric	Unit	Description
senhub.oracle_enterprise.ash.active_sessions	{session}	Active sessions over the last 5 minutes, split by wait_class
senhub.oracle_enterprise.ash.cpu_sessions	{session}	Active sessions on CPU over the last 5 minutes

RAC (gv$ views — when collect_rac: true)

Metric	Unit	Description
senhub.oracle_enterprise.rac.instances	{instance}	Number of open cluster instances
senhub.oracle_enterprise.rac.network.io	By	Cumulative SQL*Net bytes, per RAC instance (rac.instance attribute)
senhub.oracle_enterprise.rac.gc.blocks_received	{block}	Cumulative global-cache CR blocks received, per RAC instance (rac.instance attribute)

rac.network.io and rac.gc.blocks_received are cumulative counters — consumers should apply a rate() (or PRTG delta) to read per-second throughput.

Data Guard (v$dataguard_stats)

Metric	Unit	Description
senhub.oracle_enterprise.dataguard.apply_lag	s	Redo apply lag on the standby
senhub.oracle_enterprise.dataguard.transport_lag	s	Redo transport lag to the standby

Optional families

RAC metrics are only collected when collect_rac: true and the gv$ views are accessible. Data Guard metrics are only emitted when v$dataguard_stats returns rows (a standby / Data Guard configuration); a primary-only instance simply reports no Data Guard series. Each family is independent — a failure in one does not prevent the others from emitting.

Requirements

• Oracle Database Enterprise Edition with the Diagnostics Pack option licensed (required for v$sysmetric and v$active_session_history).
• A database user with CREATE SESSION and SELECT on the diagnostic views: v$sysmetric, v$active_session_history, v$version, v$instance, v$database, v$dataguard_stats, and — for RAC — gv$instance and gv$sysstat. Granting SELECT_CATALOG_ROLE (or SELECT ANY DICTIONARY) covers all of them; no administrative rights are required.
• Network path from the agent host to the database listener (default TCP port 1521).

Outputs

Oracle Enterprise metrics are available through every configured output — OTLP, Prometheus, and the pull formats (PRTG, Nagios, Web UI). For PRTG and Nagios, query the probe by its configured name:

curl "http://localhost:8080/api/{agentkey}/prtg/metrics/oracle-enterprise-prod"
curl "http://localhost:8080/api/{agentkey}/nagios/metrics/oracle-enterprise-prod"

Metric reference

Every metric this probe can emit. Metric is the OpenTelemetry name the OTLP, Prometheus and Zabbix outputs derive theirs from. Name is what a Nagios check and the API metrics= filter match. PRTG channel is the label PRTG shows, placeholders filled from the series' tags.

Metric	Name	PRTG channel	Unit	Description
senhub.oracle_enterprise.up	oracle_enterprise_up	Instance Reachable	#	1 when the Oracle instance answered this cycle, else 0
senhub.oracle_enterprise.awr.db_time	oracle_enterprise_awr_db_time	DB Time / s	s	Database time consumed per second (v$sysmetric)
senhub.oracle_enterprise.awr.db_cpu	oracle_enterprise_awr_db_cpu	DB CPU / s	s	Database CPU time consumed per second (v$sysmetric)
senhub.oracle_enterprise.awr.parse.hard	oracle_enterprise_awr_parse_hard	Hard Parses / s	#	Hard parse count per second (v$sysmetric)
senhub.oracle_enterprise.awr.parse.soft	oracle_enterprise_awr_parse_soft	Soft Parses / s	#	Soft parse count per second (v$sysmetric)
senhub.oracle_enterprise.awr.logical_reads	oracle_enterprise_awr_logical_reads	Logical Reads / s	#	Logical reads per second (v$sysmetric)
senhub.oracle_enterprise.awr.physical_reads	oracle_enterprise_awr_physical_reads	Physical Reads / s	#	Physical reads per second (v$sysmetric)
senhub.oracle_enterprise.awr.physical_writes	oracle_enterprise_awr_physical_writes	Physical Writes / s	#	Physical writes per second (v$sysmetric)
senhub.oracle_enterprise.awr.executions	oracle_enterprise_awr_executions	Executions / s	#	SQL executions per second (v$sysmetric)
senhub.oracle_enterprise.ash.active_sessions	oracle_enterprise_ash_active_sessions	Active Sessions ({wait_class})	#	Active sessions over the last 5 minutes, per wait class (ASH)
senhub.oracle_enterprise.ash.cpu_sessions	oracle_enterprise_ash_cpu_sessions	Active Sessions On CPU	#	Active sessions on CPU over the last 5 minutes (ASH)
senhub.oracle_enterprise.rac.instances	oracle_enterprise_rac_instance_count	RAC Instances	#	Number of open cluster instances visible via gv$instance
senhub.oracle_enterprise.rac.network.io	oracle_enterprise_rac_network_bytes	SQL*Net Bytes (inst {inst_id})	Bytes	Cumulative SQL*Net bytes exchanged with clients, per RAC instance
senhub.oracle_enterprise.rac.gc.blocks_received	oracle_enterprise_rac_gc_blocks_received	GC CR Blocks Received (inst {inst_id})	#	Cumulative global-cache CR blocks received, per RAC instance
senhub.oracle_enterprise.dataguard.apply_lag	oracle_enterprise_dataguard_apply_lag	Data Guard Apply Lag	s	Redo apply lag on the standby, in seconds (v$dataguard_stats)
senhub.oracle_enterprise.dataguard.transport_lag	oracle_enterprise_dataguard_transport_lag	Data Guard Transport Lag	s	Redo transport lag to the standby, in seconds (v$dataguard_stats)