Oracle Database (OracleDB) is foundational for many enterprise applications, and robust monitoring is essential for maintaining performance and reliability. KloudMate provides comprehensive visibility into OracleDB by delivering real-time insights through logs and metrics using the KloudMate Agent powered by OpenTelemetry.
OracleDB Monitoring in KloudMate enables centralized monitoring of Oracle Database instances running on AWS EC2, Azure Virtual Machines, or on-premise servers.
With OracleDB Monitoring enabled, KloudMate collects telemetry that provides visibility into:
Database session and system statistics
Tablespace utilization and data file metrics
Resource limits and consumption patterns
Performance bottlenecks and resource contention
Server logs and operational events
This visibility helps identify capacity issues, performance degradation, storage constraints, and availability problems.
Oracle Database must be installed and running
KloudMate Agent installed on the OracleDB host (see agent installation for Linux and Windows )
Monitoring database user requires:
GRANT SELECT ON V_$ SESSION TO < username > ; GRANT SELECT ON V_$SYSSTAT TO < username > ; GRANT SELECT ON V_$RESOURCE_LIMIT TO < username > ; GRANT SELECT ON DBA_TABLESPACES TO < username > ; GRANT SELECT ON DBA_DATA_FILES TO < username > ; GRANT SELECT ON DBA_TABLESPACE_USAGE_METRICS TO < username > ; Two named receiver instances are used to logically separate SQL performance queries from session queries.
Receiver Purpose sqlquery/high_cpuSQL performance queries for top CPU, top elapsed time, top executions, and deadlock logs sqlquery/sessionsActive session and wait event queries
Metric Prefix Description oracle.sql.*Top SQL by CPU time oracle.el.*Top SQL by elapsed time oracle.ex.*Top SQL by executions oracle.sess.*Active session wait events
Log in to the KloudMate platform
Go to Settings → Agents
Select the agent running on the PostgreSQL host
Click Actions → Collector Configuration
YAML editor opens for configuration
receivers : oracledb : datasource : "oracle://otel:passwd@<host-ip-address>/XEPDB1" Connection string format: oracle://username:password@host:port/service_name
Replace:
otel → monitoring usernamepasswd → monitoring user password<host-ip-address> → OracleDB host IPXEPDB1 → PDB/service name (adjust for your database)
Two named receiver instances are used to logically separate SQL performance queries from session queries.
sqlquery/high_cpu : driver : oracle host : <host-ip> database : <database-connection-string> port : 1521 collection_interval : 10s username : sys password : password storage : file_storage queries : - sql : > SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", cpu_time/1000000 AS "cpu_seconds", elapsed_time/1000000 AS "elapsed_seconds", buffer_gets AS "buffer_gets", disk_reads AS "disk_reads", ROUND(cpu_time/DECODE(executions,0,1,executions)/1000000,2) AS "cpu_per_exec" FROM v$sql ORDER BY cpu_time DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.sql.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_text - parsing_schema_name - sql_id - metric_name : oracle.sql.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.buffergets value_column : buffer_gets value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.diskreads value_column : disk_reads value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.cpuexec value_column : cpu_per_exec value_type : double data_type : gauge attribute_columns : - sql_id - parsing_schema_name - sql_text - sql : > SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", elapsed_time/1000000 AS "elapsed_seconds", cpu_time/1000000 AS "cpu_seconds", buffer_gets AS "buffer_gets", disk_reads AS "disk_reads", ROUND(elapsed_time/DECODE(executions,0,1,executions)/1000000,2) AS "elapsed_per_exec" FROM v$sql ORDER BY elapsed_time DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.el.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.buffergets value_column : buffer_gets value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.diskreads value_column : disk_reads value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.elapseexec value_column : elapsed_per_exec value_type : double data_type : gauge attribute_columns : - sql_id - sql_text - parsing_schema_name - sql : | SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", cpu_time/1000000 AS "cpu_seconds", elapsed_time/1000000 AS "elapsed_seconds" FROM v$sql WHERE executions > 0 ORDER BY executions DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.ex.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.ex.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.ex.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - sql : > SELECT originating_timestamp AS "originating_timestamp", to_char(originating_timestamp, 'DD-MON-YYYY HH24:MI:SS') AS "time", message_text AS "message_text" FROM v$diag_alert_ext WHERE message_text LIKE '%deadlock%' ORDER BY originating_timestamp DESC tracking_column : originating_timestamp tracking_start_value : 2024-01-01T00:00:00Z logs : - body_column : message_text attribute_columns : - time - originating_timestamp sqlquery/sessions : driver : oracle host : <host-ip> database : <database-connection-string> port : 1521 collection_interval : 30s username : sys password : password storage : file_storage queries : - sql : | SELECT s.sid AS "sid", s.username AS "username", s.status AS "status", s.osuser AS "osuser", s.machine AS "machine", s.program AS "program", s.sql_id AS "sql_id", s.event AS "event", s.wait_class AS "wait_class", s.seconds_in_wait AS "seconds_in_wait", s.state AS "state" FROM v$session s WHERE s.status = 'ACTIVE' AND s.username IS NOT NULL ORDER BY s.seconds_in_wait DESC metrics : - metric_name : oracle.sess.secinwait value_column : seconds_in_wait value_type : int data_type : gauge attribute_columns : - sid - username - machine - program - sql_id - event - wait_class - state - status exporters : otlphttp : endpoint : 'https://otel.kloudmate.com:4318' headers : Authorization : <API-Key> This enables forwarding telemetry to KloudMate for analysis and visualization.
extensions : file_storage : create_directory : true receivers : oracledb : datasource : "oracle://otel:passwd@<host-ip-address>/XEPDB1" sqlquery/high_cpu : driver : oracle host : <server_ip_address> database : XEPDB1 port : 1521 collection_interval : 10s username : sys password : <password> storage : file_storage queries : - sql : > SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", cpu_time/1000000 AS "cpu_seconds", elapsed_time/1000000 AS "elapsed_seconds", buffer_gets AS "buffer_gets", disk_reads AS "disk_reads", ROUND(cpu_time/DECODE(executions,0,1,executions)/1000000,2) AS "cpu_per_exec" FROM v$sql ORDER BY cpu_time DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.sql.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_text - parsing_schema_name - sql_id - metric_name : oracle.sql.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.buffergets value_column : buffer_gets value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.diskreads value_column : disk_reads value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - parsing_schema_name - sql_text - metric_name : oracle.sql.cpuexec value_column : cpu_per_exec value_type : double data_type : gauge attribute_columns : - sql_id - parsing_schema_name - sql_text - sql : > SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", elapsed_time/1000000 AS "elapsed_seconds", cpu_time/1000000 AS "cpu_seconds", buffer_gets AS "buffer_gets", disk_reads AS "disk_reads", ROUND(elapsed_time/DECODE(executions,0,1,executions)/1000000,2) AS "elapsed_per_exec" FROM v$sql ORDER BY elapsed_time DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.el.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.buffergets value_column : buffer_gets value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.diskreads value_column : disk_reads value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.el.elapseexec value_column : elapsed_per_exec value_type : double data_type : gauge attribute_columns : - sql_id - sql_text - parsing_schema_name - sql : | SELECT * FROM ( SELECT sql_id AS "sql_id", sql_text AS "sql_text", parsing_schema_name AS "parsing_schema_name", executions AS "executions", cpu_time/1000000 AS "cpu_seconds", elapsed_time/1000000 AS "elapsed_seconds" FROM v$sql WHERE executions > 0 ORDER BY executions DESC ) WHERE ROWNUM <= 10 metrics : - metric_name : oracle.ex.executions value_column : executions value_type : int data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.ex.cpuseconds value_column : cpu_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - metric_name : oracle.ex.elapseseconds value_column : elapsed_seconds value_type : double data_type : sum aggregation : cumulative monotonic : true attribute_columns : - sql_id - sql_text - parsing_schema_name - sql : > SELECT originating_timestamp AS "originating_timestamp", to_char(originating_timestamp, 'DD-MON-YYYY HH24:MI:SS') AS "time", message_text AS "message_text" FROM v$diag_alert_ext WHERE message_text LIKE '%deadlock%' ORDER BY originating_timestamp DESC tracking_column : originating_timestamp tracking_start_value : 2024-01-01T00:00:00Z logs : - body_column : message_text attribute_columns : - time - originating_timestamp sqlquery/sessions : driver : oracle host : <server_ip_address> database : XEPDB1 port : 1521 collection_interval : 30s username : sys password : password storage : file_storage queries : - sql : | SELECT s.sid AS "sid", s.username AS "username", s.status AS "status", s.osuser AS "osuser", s.machine AS "machine", s.program AS "program", s.sql_id AS "sql_id", s.event AS "event", s.wait_class AS "wait_class", s.seconds_in_wait AS "seconds_in_wait", s.state AS "state" FROM v$session s WHERE s.status = 'ACTIVE' AND s.username IS NOT NULL ORDER BY s.seconds_in_wait DESC metrics : - metric_name : oracle.sess.secinwait value_column : seconds_in_wait value_type : int data_type : gauge attribute_columns : - sid - username - machine - program - sql_id - event - wait_class - state - status processors : memory_limiter : check_interval : 1s limit_mib : 1000 spike_limit_mib : 800 batch : send_batch_size : 1000 timeout : 10s resource : attributes : - key : service.name action : insert from_attribute : host.name resourcedetection : detectors : - system system : resource_attributes : host.name : enabled : true host.id : enabled : true os.type : enabled : false exporters : debug : verbosity : detailed otlphttp : endpoint : 'https://otel.kloudmate.com:4318' headers : Authorization : <API-Key> service : pipelines : metrics : receivers : [ oracledb , sqlquery/high_cpu , sqlquery/sessions ] processors : [ resourcedetection , resource , memory_limiter , batch ] exporters : [ debug , otlphttp ] logs : receivers : [ oracledb , sqlquery/high_cpu ] processors : [ resourcedetection , resource , memory_limiter , batch ] exporters : [ debug , otlphttp ] extensions : - file_storage Verify that metrics are flowing into KloudMate using the Explore view.
After the agent restarts:
Log in to KloudMate
Navigate to Explore
Select OpenTelemetry → Metrics
Choose a OracleDB metric and run the query
Seeing time-series data confirms that OracleDB telemetry is flowing successfully.
Access KloudMate and create dashboards to visualize OracleDB metrics and logs. Configure alerting rules for critical thresholds and unusual activity.
The following metrics are collected through the oracledb receiver:
Default Metrics oracledb_cpu_time oracledb_dml_locks_limit oraclb_dml_locks_usage oracledb_enqueue_deadlocks oracledb_enqueue_locks_limit oracledb_enqueue_locks_usage oracledb_enqueue_resources_limit oracledb_enqueue_resources_usage oracledb_exchange_deadlocks oracledb_executions oracledb_hard_parses Oracledb_logical_reads oracledb_parse_calls oracledb_pga_memory oracledb_physical_reads oracledb_processes_limit oracledb_processes_usage oracledb_sessions_limit oracledb_sessions_usage oracledb_tablespace_size_limit Oracledb_tablespace_size_usage oracledb_transactions_limit oracledb_transactions_usage oracledb_user_commits oracledb_user_rollbacks
Optional Metrics oracledb_consistent_gets oracledb_db_block_gets
For the complete metrics list, refer to the metrics reference .
