OracleDB Monitoring - performance and table space stats

OracleDB Monitoring - performance and table space stats

Use oracledb-exporter to monitor oracle database. Compared to other dashboards, tablespace monitoring is added, which can be expanded to set tablespace alarms;

Credits:

Based on https://grafana.com/grafana/dashboards/11121

Note: I translated chinese labels to english and fixed tablespace usage stats which was broken since grafana 7.0 (grafana removed ’timeseries_to_rows’ in v7.0 which was fixed by applying following transformations:

1. “Label to fields” (Leave value field option empty)
2. “Group by” (Set ’tablespace’ to ‘Group by’ and value to ‘Calculate’ (Last (not null)))

Prerequisites: oracle db, prometheus, oracledb-exporter

oracledb-exporter https://github.com/iamseth/oracledb_exporter

Running oracledb-exporter

• Please check documentation at https://github.com/iamseth/oracledb_exporter

Ensure that the environment variable DATA_SOURCE_NAME is set correctly before starting.

Since oracledb-exporter relies on oracle instantclient, you might also need to set LD_LIBRARY_PATH environment variable to instantclient location, i.e.:

export LD_LIBRARY_PATH=/home/DEV/plugins/instantclient_21_1:$LD_LIBRARY_PATH

(if you get errors like oracledb_exporter: error while loading shared libraries: libclntsh.so)

There is also a docker image which provides oracledb-exporter: https://hub.docker.com/r/iamseth/oracledb_exporter

To run docker image:

docker pull iamseth/oracledb_exporter

docker run -d –name oracledb_exporter –restart=always -p 9161:9161 -e DATA_SOURCE_NAME=[db_user]/[db_passwd]@[db_ip]:[db_port]/[SID] iamseth/oracledb_exporter

Check http://localhost:9161/metrics after running to verify metrics are up

Prometheus config to scrape logs from oracledb-exporter

scrape_configs:

• job_name: ‘oracle_metrics’ scrape_interval: 5s static_configs:
  • targets: [’localhost:9161’]