GreenPlum¶

Greenplum is a high-performance, highly available database system based on the Massively Parallel Processing (MPP) architecture. It is mainly used for processing and analyzing large-scale datasets. Greenplum applies to data warehousing, business intelligence, and big data analytics scenarios, especially when handling PB-level data volumes, providing efficient data storage and analysis capabilities. The observability of Greenplum includes monitoring database performance, detecting and notifying failures, as well as visualizing the system's operational status.

Installation Configuration¶

Prerequisites¶

Greenplum version 5 or 6, tested versions:

• 6.24.3

Exporter Installation¶

You can download and install greenplum exporter, which is a data collector that exposes Greenplum metrics as Prometheus data and supports both GP5 and GP6.

• For example, on CentOS, the startup command is as follows:

export GPDB_DATA_SOURCE_URL=postgres://gpadmin:[email protected]:5432/postgres?sslmode=disable
./greenplum_exporter --web.listen-address="0.0.0.0:9297" --web.telemetry-path="/metrics" --log.level=error

• If running via Docker, the startup command is as follows:

docker run -d -p 9297:9297 -e GPDB_DATA_SOURCE_URL=postgres://gpadmin:[email protected]:5432/postgres?sslmode=disable inrgihc/greenplum-exporter:latest

Note: The environment variable GPDB_DATA_SOURCE_URL specifies the connection string for connecting to the Greenplum database (please use the gpadmin account to connect to the postgres database). This connection string starts with postgres://, and its format is as follows:

postgres://gpadmin:[email protected]:5432/postgres?sslmode=disable
postgres://[Database connection account, must be gpadmin]:[Account password, i.e., gpadmin's password]@[Database IP address]:[Database port]/[Database name, must be postgres]?[Parameter name]=[Parameter value]&[Parameter name]=[Parameter value]

Then access the URL for monitoring metrics: http://127.0.0.1:9297/metrics

More startup parameters:

usage: greenplum_exporter [<flags>]

Flags:
  -h, --help                   Show context-sensitive help (also try --help-long and --help-man).
      --web.listen-address="0.0.0.0:9297"
                               web endpoint
      --web.telemetry-path="/metrics"
                               Path under which to expose metrics.
      --disableDefaultMetrics  do not report default metric
      (go metrics and process metrics)
      --version                Show application version.
      --log.level="info"       Only log messages with the given severity or above. Valid levels: [debug, info, warn, error, fatal]
      --log.format="logger:stderr"
                               Set the log target and format. Example: "logger:syslog?appname=bob&local=7" or "logger:stdout?json=true"

Collector Configuration¶

Host Installation¶

• Install Datakit

Configure Collector¶

Since greenplum exporter can directly expose metrics URLs, it can be collected directly through the prom collector by configuring the corresponding URL for collection. Navigate to the conf.d/prom directory under the DataKit installation directory, copy prom.conf.sample and rename it to greenplum.conf

cp prom.conf.sample greenplum.conf

Adjust the content of greenplum.conf as follows:

[[inputs.prom]]
  ## Exporter URLs.
  urls = ["http://127.0.0.1:9297/metrics"]
  ## Collector alias.
  source = "greenplum"
...

Other configurations should be adjusted as needed, Parameter adjustment description:

• urls: prom metrics address; here fill in the metrics url exposed by greenplum. If DataKit is installed on the Greenplum host, the default is http://127.0.0.1:9297/metrics
• source: data source, change to greenplum for easier filtering of metrics later
• interval: collection interval

Restart Datakit¶

Restart Datakit

Metrics¶

GreenPlum Metric Sets¶