overcomplicated setup and open questions

[wuast94]

I managing a swarm cluster and have dealed with many wired applications to setup, but zabbix seems to be special.

db init containers are kinda bad, it works for a compose file but in a swarm cluster depends on for example is not supporting and running a one shot container is also a thing. all projects i know had integraded that into the main application, check if db exists if not init and go.

the tags are wired to, the latest is is 7.2.2, basic semver nice. but the zabbix-server-pgsql for example has 7.2.2-ubuntu that was pushed 19 days ago and a 7.2-ubuntu-latest tag that is newer and pushed 8 days ago, that makes no sense, why a semver when there is some magic half semver with latest?

a thousand volume mappings, no big deal and maybe related to historical zabbix setups migrated to docker.

you mentioning profiles but to look which profiles are available we have to look into each docker compose file and i didnt even understand what i should use and which component i need for what like selenium or elsaticsearch, something that could be find in docs anywhere but would be nice to get an overview.

it feels like i have to setip a vm or special host to get this up and running, and at that point its kinda yolo if using docker source or packages or some install scripts or whatever.

all the config files that needs to be copyed etc could be autocreated by the container if not existing.

docker and also this project i think have in common to streamline and simply things, but this repo overcomplicates many things in my opinion and throws more questions than it solves.

this isnt meant as hate or anything like that. i just would love to play around with zabbix and maybe use it for my homelab, but its not easy to understand all the things.

[dotneft]

all projects i know had integraded that into the main application, check if db exists if not init and go.

It is included, do not need to use run "special" container for that, server and proxy do it by default. All compose files are examples, how to do it. And include redundant steps, configuration.

why a semver when there is some magic half semver with latest?

Because static tags are releases, while "latest" can include changes in Docker files, but do not include changes in Zabbix itself at all.

a thousand volume mappings, no big deal and maybe related to historical zabbix setups migrated to docker.

Zabbix has many possibilities and all of them are added by default, if you do not need, just do not use these volumes

you mentioning profiles but to look which profiles are available we have to look into each docker compose file and i didnt even understand what i should use and which component i need for what like selenium or elsaticsearch, something that could be find in docs anywhere but would be nice to get an overview.

use compose config instead.

all the config files that needs to be copyed etc could be autocreated by the container if not existing.

can be, but since Zabbix 7.2 we started to use env variables in configuration files, that is why such configs added to the project.

docker and also this project i think have in common to streamline and simply things, but this repo overcomplicates many things in my opinion and throws more questions than it solves.

usually you do not need to care about dockerfiles, you just use it or extend according your requirements. That is why it looks like a bit complicated, but for usage and extend purposes they are good.