Reverse engineering on the Zabbix DB

[Fof]

Hi Zbx community,

I am trying to dig into the Zabbix DB in order to use some data gathered by zabbix server. Actually, I'd like to do some stuff outside the Zabbix'frontend like :
- Use the Zabbix user management
- Be able to link a user with his hosts
- whatever...

For this purpose, I'm trying to build a data conceptual model from the DB structure file (I think there already are some tools which are able to do the job) and I am trying to gather any information about the Zabbix DB.

An old DB structure explaination with information on what tables are used to (Zabbix 1.0), p114:
http://www.zabbix.com/Zabbix%20Manual.pdf

Here are some information who could help someone to perform something as I am trying to do but I am very interested by any information about the Zabbix DB.

The following schema is my current draft.

Cheers.

[wrocha]

Great job

Nice working. It is helping me a lot.

[wrocha]

Zabbix 2.2 Model

I used "Mysql Workbench" to create this model. I did supressed some tables but still very usefull.

[michaelwemoto]

Hi,

Here is the postgresql schema, I found this post and decided I would submit mine for you.

Courtesy of SchemaSpy

[wrocha]

Thank you michaelwemoto! that is really helpfull!

[Linwood]

If you haven't already noticed, the API reference at the end of the manual has (more or less) a complete documentation of most fields in key tables.

What I've found difficult in prowling the database manually is figuring out where the specializations are, e.g. a host is also a template, but beyond that how (for example) items that are templated are different from ones that are not, and are different still from items that are from discovery, and are different still from items from discovery from a template. It's kind of magic at the API level and not all that obvious in SQL, you just have to experiment with each such structure.