Good morning Team,
I am currently using Traccar v6.10.0 with a MySQL database in my setup. Traccar is responsible for processing device data and forwarding position updates to Redis via its WebSocket API, which are then consumed by custom frontend applications.
My goal is to fully decouple Traccar from the rest of the system and dedicate it solely to data processing. While the current setup is functional, I still rely on Traccar for generating reports (trips).
The main reason for decoupling is to keep the Traccar database as lean as possible. I plan to store trips in my own database, where clients can access them directly. Traccar will only store position data for up to one month, as my devices transmit GPRS signals at 20-30 second intervals, generating a large volume of data.
As part of my experiment, I tried using Traccar's WS socket Events (Device Moving and Device Stop) to calculate trips. However, I found this approach to be unreliable, as it doesn't consistently track trips and result in inaccuracies.
Any recommendations or insights on how best to implement this architecture would be greatly appreciated.
Thanks.
Good morning Team,
I am currently using Traccar v6.10.0 with a MySQL database in my setup. Traccar is responsible for processing device data and forwarding position updates to Redis via its WebSocket API, which are then consumed by custom frontend applications.
My goal is to fully decouple Traccar from the rest of the system and dedicate it solely to data processing. While the current setup is functional, I still rely on Traccar for generating reports (trips).
The main reason for decoupling is to keep the Traccar database as lean as possible. I plan to store trips in my own database, where clients can access them directly. Traccar will only store position data for up to one month, as my devices transmit GPRS signals at 20-30 second intervals, generating a large volume of data.
As part of my experiment, I tried using Traccar's WS socket Events (Device Moving and Device Stop) to calculate trips. However, I found this approach to be unreliable, as it doesn't consistently track trips and result in inaccuracies.
Any recommendations or insights on how best to implement this architecture would be greatly appreciated.
Thanks.