I'm new to zabbix and trying to build a decent zabbix instance.
Please correct me where I am wrong or not fully understanding something.
I've been researching a bit and it sounds like postgresql with timescaledb may be a better performing option than mysql/mariadb. Then I hear about partitioning the tables in the database to increase performance and make housekeeping faster.
I've stumbled across this project and I'm wondering if it is a valid solution for zabbix 5.x?
https://github.com/Doctorbal/zabbix-...s-partitioning
So right now I'm planning on a Debian server with zabbix 5, postgresql with timescaledb extension installed and nginx.
The environment being monitored isn't huge, maybe under 200 virtual servers (mostly windows), under 20 hypervisors.. a bunch of network equipment.. switches, shared storage etc..
Am on on the right track here? Does it make sense to partition the postgresql database tables right from the start rather than try to manage it all later when a problem surfaces?
Please correct me where I am wrong or not fully understanding something.
I've been researching a bit and it sounds like postgresql with timescaledb may be a better performing option than mysql/mariadb. Then I hear about partitioning the tables in the database to increase performance and make housekeeping faster.
I've stumbled across this project and I'm wondering if it is a valid solution for zabbix 5.x?
https://github.com/Doctorbal/zabbix-...s-partitioning
So right now I'm planning on a Debian server with zabbix 5, postgresql with timescaledb extension installed and nginx.
The environment being monitored isn't huge, maybe under 200 virtual servers (mostly windows), under 20 hypervisors.. a bunch of network equipment.. switches, shared storage etc..
Am on on the right track here? Does it make sense to partition the postgresql database tables right from the start rather than try to manage it all later when a problem surfaces?
Comment