Each script is purposefully built self-contained entities.
All python scipts must be invoked from the virtualenv prepared from requirements file  

All contents of this repo is made to make zabbix administration easier. Tasks such as database backup, partitioning mysql 
is scripted out and run from cron only. The little intelligence built into the scripts are made to withstand some visible bugs, if not all.
The scripts are run in my production environment, where:
Total DB size: ~800GB with 3 month retention
Total applications integrated: 12
Total Linux/UNIX hosts: 250+
And many custom items.

partition_mgt.py
This script is meant to partition the mysql database for zabbix into user's own choice of number of partitions on a future month.
The partitioning scheme is range only and no subpartition is used. There's explicit use to data directory in partition, because to split out the IO across the LV. My production is meant to have pv-vg-lv as one to one relation, to distribute IO to maximum level from OS.
However, there are more tunings done on mysql size which is in my.cnf


mysql-backup.sh
This script actually dumps tables in two stages.
1. without data for all tables 
2. with data for specific tables which are less than 1200MB in size. 

Both stages dump in two different files. One skeletal with table definition, another with data (usually this one is configs).
An rsync is run to sync all physical data files. This is the 3rd measure to dump entire DB. This is error prone and requires longer time to recover database.