73 lines
2.3 KiB
ReStructuredText
73 lines
2.3 KiB
ReStructuredText
Monitoring
|
|
==========
|
|
|
|
Notfellchen should, like every other software, be easy to monitor. Therefore a basic metrics are exposed to `https://notfellchen.org/metrics`.
|
|
The data is encoded in JSON format and is therefore suitable to bea read by humans and it is easy to use it as data source for further processing.
|
|
|
|
|
|
Exposed Metrics
|
|
---------------
|
|
|
|
.. code::
|
|
|
|
users: number of users (all roles combined)
|
|
staff: number of users with staff status
|
|
adoption_notices: number of adoption notices
|
|
adoption_notices_by_status: number of adoption notices by major status
|
|
adoption_notices_without_location: number of location notices that are not geocoded
|
|
|
|
Example workflow
|
|
----------------
|
|
|
|
To use the exposed metrics you will usually need a time series database and a visualization tool.
|
|
|
|
As time series database we will utilize InfluxDB, the visualization tool will be Grafana.
|
|
|
|
InfluxDB and Telegraf
|
|
^^^^^^^^^^^^^^^^^^^^^
|
|
|
|
First we install InfluxDB (e.g. with docker, be aware of the security risks!).
|
|
|
|
.. code::
|
|
|
|
# Pull the image
|
|
$ sudo docker pull influxdb
|
|
|
|
# Start influxdb
|
|
$ sudo docker run -d -p 8086:8086 -v influxdb:/var/lib/influxdb --name influxdb influxdb
|
|
|
|
# Start influxdb console
|
|
$ docker exec -it influxdb influx
|
|
Connected to http://localhost:8086 version 1.8.3
|
|
InfluxDB shell version: 1.8.3
|
|
> create database monitoring
|
|
> create user "telegraf" with password 'mypassword'
|
|
> grant all on monitoring to telegraf
|
|
|
|
.. note::
|
|
When creating the user telegraf check the double and single quotes for username an password.
|
|
|
|
Now install telegraf and configure `etc/telegraf/telegraf.conf`. Modify the domain and your password for the InfluxDB database.
|
|
|
|
.. literalinclude:: example.telegraf.conf
|
|
:linenos:
|
|
:language: python
|
|
|
|
Graphana
|
|
^^^^^^^^
|
|
|
|
Now we can simply use the InfluxDB as data source in Grafana and configure until you have
|
|
beautiful plots!
|
|
|
|
.. image:: monitoring_grafana.png
|
|
|
|
Healthchecks
|
|
------------
|
|
|
|
You can configure notfellchen to give a hourly ping to a healthchecks server. If this ping is not received, you will get notified and cna check why the celery jobs are no running.
|
|
Add the following to your `notfellchen.cfg` and adjust the URL to match your check.
|
|
.. code::
|
|
|
|
[monitoring]
|
|
healthchecks_url=https://health.example.org/ping/5fa7c9b2-753a-4cb3-bcc9-f982f5bc68e8
|