Skip to content

Latest commit

 

History

History
94 lines (69 loc) · 3.05 KB

README.md

File metadata and controls

94 lines (69 loc) · 3.05 KB
Raw

Introduction

This demo shows how to use influxdb and grafana to show metrics. There are two approaches:

  • Populate the database with data
  • Use telegraf to get the metrics

Both are complementary.

Case A: The first one is useful to prepopulate the database with data that is not monitored by telegraf. Historic data. Then use telegraf to get the metrics regularly.

Case B: The first one is useful to show metrics from a database that is not monitored by telegraf. Manually will be populated with new data. The second one is useful to show metrics from a database that is monitored by telegraf.

Requeriments

  • podman

Installation and configuration

Whit the script launch.sh you can start a podman pod and the necessary containers.

./launch.sh

Check your containers using

podman ps

These containers exports the port 8086 for influxdb and 3000 for grafana.

Configure influxDB

First is necessary to configure the database access. And get the token to access the database from the grafana.

  1. Access to the web http://localhost:8086
  2. Fill the info. By default for the demo use metrics as organization and as a name for the bucket.
  3. Copy the token

Configure grafana

Secondly is necessary to configure grafana to access the database.

  1. Access to the web page at http://localhost:3000 (user: admin, pass: admin)
  2. Add a new data source:
  3. select influxdb
  4. select query language: flux
  5. set the url for the DB to http://influxdb:8086
  6. set the organization to metrics
  7. set the token
  8. set the default bucket to metrics

Populate the database

Option 1: From pre-generated data

There is a data file pre-generatred that you can use. But you can generate your own data (see the section for dump data from progress/bugzilla and generate metrics)

Load the data. Execute the following commands: Note: replace with the token generated in the previous section. Also ensure the bucket and organization are correct.

podman cp data.txt influxdb:/tmp
podman exec -ti influxdb influx write -b metrics -o metrics -t <token>  -p ns --format=lp -f /tmp/data.txt

Try it on grafana creating a new dashboard and a new panel. Create a query using flux. E.g.:

from(bucket: "metrics")
  |> range(start: -20d)
  |> filter(fn: (r) => r["_measurement"] == "metrics")
  |> filter(fn: (r) => r["_field"] == "IN_PROGRESS")
  |> aggregateWindow(every: v.windowPeriod, fn: mean, createEmpty: false)
  |> yield(name: "mean")

Option 2: Use telegraf

  1. Copy your database (obtained with the dumpper) to the demo directory with the name database.db
  2. Copy telegraf_template to telegraf.conf
  3. Modify the telegraf.com to specify the correct token
  4. Execute the following command:
./launch_telegraf.sh

This will execute all the necessary commands to allow the execution of the script metrics_snapshot.sh Note: In the configuration of telegraf, the first execution for metrics_snapshot.sh will start 2 minutes after the start of telegraf. If the inicialization proces didn't finish in 2 minutes, the first execution will fail. But the next executions will work.