NOTE: VAST versions 4.5 and later include the Prometheus Exporter as part of the official VMS API and it should be used instead of this external exporter. More information is available on the VAST support site for versions 4.6, 4.7, 5.0, and 5.1.

---

VAST Prometheus Exporter

The Prometheus exporter connects to VMS and leverages its REST API to extract state and metric information. It listens to port 8000 by default. This can be changed using the --port parameter.

Pre-requisities

VAST versions 4.2 to 4.4.

VAST versions >= 4.5 should use the exporter included with VMS.

Metrics Provided

1. Cluster capacity and states.
2. Physical component states (NIC, node, drive, etc').
3. Logical objects: view, quota, replication targets, etc'.
4. Performance metrics (BW/IOPS/latency/etc') for cluster, nodes, views and top users.

Docker Installation

# build a local docker image
$ ./build.sh
# run a docker container in the background
$ ./run.sh --user=<USER> --password=<PASSWORD> --address=<ADDRESS>

Python Installation

$ pip install -r requirements.txt
$ ./vast_exporter.py --user=<USER> --password=<PASSWORD> --address=<ADDRESS>

Usage

Beyond the user/password/address parameters which are required there are optional parameters. Optional features inlude specifying a custom SSL certificate and collection of top user performance data. Run the exporter with --help to see a description.

Testing

To run the exporter once and make sure no errors are raised run the following:

$ ./vast_exporter.py --user=<USER> --password=<PASSWORD> --address=<ADDRESS> --test
2022-04-28 11:36:47,045 MainThread INFO: VAST Exporter started running. Listening on port 8001
2022-04-28 11:36:58,658 MainThread INFO: Collection is successful!

To run the exporter forever remove the --test parameter and get the output using curl:

$ curl http://localhost:8000 | grep vast_collector
# HELP vast_collector_latency Total collection time
# TYPE vast_collector_latency summary
vast_collector_latency_count 1.0
vast_collector_latency_sum 7.011876339
...

Monitoring

Besides the obious stdout output where errors will land, the following metric will be incremeneted upon an error:

vast_collector_errors_total

Prometheus Config

Full explanation of Prometheus configuration here: https://prometheus.io/docs/prometheus/latest/configuration/configuration/

The following snippet shows how to add the VAST exporter to Prometheus:

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: "vast"

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.
    scrape_interval: 1m
    scrape_timeout: 50s
    
    static_configs:
      - targets: ["<EXPORTER HOST>:8000"]

How to run Prometheus using docker:

docker run -p 9090:9090 -v /path/to/prometheus.yml:/etc/prometheus/prometheus.yml prom/prometheus