Skip to content
This repository has been archived by the owner on May 29, 2024. It is now read-only.

Proxy to import OpenStack Aodh alarms into Sensu Core Server

License

Notifications You must be signed in to change notification settings

josecastillolema/aodh2sensu

Repository files navigation

GitHub GitHub language count GitHub top language GitHub Pipenv locked Python version GitHub Pipenv locked dependency version GitHub Pipenv locked dependency version GitHub Workflow Status Python application CI

aodh2sensu

Imports OpenStack Aodh alarms into Sensu Core Server.

Install

Via requirements file

$ pip3 install -r ./requirements.txt

Via pipenv pipfile

$ pipenv install

Use

Usage: aodh2sensu.py [options]

Imports OpenStack Aodh alarms into Sensu Core Server.

Options:
  -h, --help            show this help message and exit
  --sensu-url URL       if not specified, defaults to localhost:4567
  • Run the aodh2sensu proxy. sensu_url must point to the sensu server. The proxy must be run in a server reachable from OpenStack controllers and with access to the Sensu Server (for example, the Sensu Server itself).

    $ ./aodh2sensu.py
    

    or alternatively via pipenv:

    $ pipenv shell
    $ ./aodh2sensu.py
    
  • Create an Aodh alarm from OpenStack side. This example alarm will trigger an HTTP POST message to the aodh2sensu proxy whenever the cpu utilization of $INSTANCE_ID goes above 20%:

    $ openstack alarm create \
    --name cpu_hi4 \
    --type gnocchi_resources_threshold \
    --description 'CPU High Average' \
    --metric cpu_util \
    --threshold 20.0 \
    --comparison-operator gt \
    --aggregation-method mean \
    --granularity 300 \
    --evaluation-periods 1 \
    --resource-type instance \
    --resource-id $INSTANCE_ID \
    --alarm-action 'http://x.y.z.w:50000' \
    --ok-action 'http://x.y.z.w:50000' \
    --insufficient-data-action 'http://x.y.z.w:50000'
    

    where x.y.z.w is the IP address of the server running aodh2sensu proxy.

  • Confirm the alarm transitions from insufficient_data state to ok state:

    $ openstack alarm list
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | alarm_id                             | type                                       | name           | state             | severity | enabled |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | c466d832-cfce-4488-9726-c631800a36b1 | gnocchi_resources_threshold                | cpu_hi4        | ok                | low      | True    |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    
  • Generate load in the instance above the 20% threeshold, and wait for the alarm to transitition to alarm state:

    $ openstack alarm list
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | alarm_id                             | type                                       | name           | state             | severity | enabled |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | c466d832-cfce-4488-9726-c631800a36b1 | gnocchi_resources_threshold                | cpu_hi4        | alarm             | low      | True    |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    
  • Check the alarm from the Uchiwa dashboard: Screenshot 1 of the Uchiwa dashboard Screenshoto 2 of the Uchiwa dashboard

  • Stop the load generation in the instance, wait for the alarm to transition back to ok state:

    $ openstack alarm list
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | alarm_id                             | type                                       | name           | state             | severity | enabled |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    | c466d832-cfce-4488-9726-c631800a36b1 | gnocchi_resources_threshold                | cpu_hi4        | ok                | low      | True    |
    +--------------------------------------+--------------------------------------------+----------------+-------------------+----------+---------+
    
  • Check the new state of the sensu alert: Screenshot 3 of the Uchiwa dashboard

  • Confirm it has disappeared from the list of active alerts: Screenshot 4 of the Uchiwa dashboard

Docker

To build the image:

$ buildah build-using-dockerfile -t aodh2sensu .

To run the image:

$ podman run --name aodh2sensu --add-host="localhost:10.88.0.1" -d -p 50000:50000 aodh2sensu

Pointing to a specific Sensu Server URL:

$ podman run --name aodh2sensu --add-host="localhost:10.88.0.1" -d -p 50000:50000 -e SENSU_URL=x.y.z.w:p aodh2sensu

Check logs:

$ podman logs aodh2sensu

Systemd

To run aodh2sensu as a systemd service.

Normal

# cat /usr/lib/systemd/system/aodh2sensu.service
[Unit]
Description=aodh2sensu service

[Service]
Type=simple
ExecStart=/opt/aodh2sensu/aodh2sensu.py

[Install]
WantedBy=multi-user.target

Docker

# cat ~/.config/systemd/user/aodh2sensu.service 
[Unit]
Description=aodh2sensu service via podman

[Service]
Restart=on-failure
ExecStartPre=/usr/bin/rm -f /%t/%n-pid /%t/%n-cid
ExecStart=/usr/bin/podman run --conmon-pidfile  /%t/%n-pid  --cidfile /%t/%n-cid --add-host="localhost:10.88.0.1" -d --name aodh2sensu -p 50000:50000 aodh2sensu
ExecStop=/usr/bin/sh -c "/usr/bin/podman rm -f `cat /%t/%n-cid`"
KillMode=none
Type=forking
PIDFile=/%t/%n-pid

[Install]
WantedBy=multi-user.target

About

Proxy to import OpenStack Aodh alarms into Sensu Core Server

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks