Introduction

This project makes use of the Zabbix API to detect problems within Zabbix monitoring and flash lights configured within Home Assistant based on the severity of whether the problems are resolved.

There is active hours which default to be between 09:00 and 23:00 so if a problem is raised outside of these errors, the lights don't flash so you don't get woken up :)

This is a python script so can run directly, but can also be run via a docker container for simple deployment.

Note there are a couple of headers CF-Access-Client-Id and CF-Access-Client-Secret which are used to authenticate to the Zabbix server hosted by Cloudflare Zero Trust. Configuring cloudflare zero trust is out of scope for this project, but if you are not using it, these headers would be ignored, or you can remove them.

However, if not using them, you may want to amend the script to not check whether the cf-* config values are not set otherwise the script won't run.

This project is also not deployed docker hub as its too custom and specific to my setup, however, you can deploy to your own local docker repository and tweak to set up for your needs.

Configuration

Config Name	Description	Default Value	Required
cf_client_id	The cloudflare zero trust client id	""	true
cf_client_secret	The cloudflare zero trust secret	""	true
cycle_time	How long to wait between each API request to Zabbix	60	false
ha_auth_token	The authentication token generated within Home Assistant	true
ha_base_url	The URL to home assistant URL <IP_ADDR>:8123/api	""	true
zabbi_api_key	The API key to send API requests to Zabbix	""	true
zabbix_url	The URL to the Zabbix API <IP ADDR>/zabbix/api_jsonrpc.php	true
alert_active_hours_only	Only flash the lights during awake hours (i.e. don't flash at early morning etc	true	false
alert_start_active_hours	The start time the lights should be able to flash	"09:00"	false
alert_end_active_hours	The end time of when the lights should stop being flashed	"23:00"	false

Notes

There's a couple of things to note:

• The home assistant URL, can technically be a DNS name, however, I noticed API requests were significantly slower when using the hostname (3 seconds) even though DNS lookups were sub second, so used the IP address.
• In the code you will notice under the Zabbi manager, there's a specific problem being ignored to not get added to the list of active problems. This problem is Buffer pool utilization is too low (less than 50% for 5m). Although this doesn't show as an active problem on the GUI and the trigger is disabled, it always gets returned.
• You will probably want to change the code in ZabbixManager.py on line 30 as that is filtering a specific zabbix group id which probably won't apply to your own zabbix set up.

Deployment

Running python script direct

Ensure that environment variables are set on your system and run the command pip3 install --no-cache-dir -r requirements.txt to install requirements.

Then run the service python3 main.py to run the script. This can be put into a systemd service on Linux for example to keep running so its always monitored.

Deploying to local docker repository

Ensure docker is running on the machine you are building from, and/or where you want to run the container form.

Run the below command to build the container

docker build --platform linux/amd64 -t zabbix-to-ha:latest .

Note you can omit --platform if you are already on x86/64 CPU architecture, however if you are building locally and pushing somewhere else, that has different architecture, such as building from M Chip Apple Silicon but deploying to a standard linux server.

Run the below command to tag and push the container to a custom repository

docker tag zabbix-to-ha:latest repository/zabbix-to-ha
docker push repository/zabbix-to-ha

Change repository for the IP:PORT if hosting on a custom docker registry.

docker run  --restart always --network bridge --name zabbix-to-ha repository/zabbix-to-ha:latest

Remember to hae the environment variables set, this can be done using the -e argument in the docker run command. For example docker run ... -e ha_base_url='http://localhost:8123/api'

To simplify things, using a docker management tool such as portainer can help deploy the docker container issue

Change repository for the IP:PORT if hosting on a custom docker registry.