Skip to content

Latest commit

 

History

History
29 lines (23 loc) · 2.02 KB

README.md

File metadata and controls

29 lines (23 loc) · 2.02 KB
Raw

external-sensor-logger

Log sensor data from a Linux machine on an external server.

What would you use this for?

In a situation where your computer is regularly shutting down to a suspected power or thermal issue, these scripts can be used to log the sensor data externally to be looked at later.

You normally would not be able to log this data reliably on a local drive as a sudden power failure can prevent the data from being written and even cause filesystem corruption.

How to use?

  • Run sensor_logger_server.py on a local server, like a Raspberry Pi.
  • Make sure the client machine has lm_sensors and curl packages installed (lm_sensors on Arch Linux, lm-sensors for Debian).
  • Run sensor_logger_client.sh with the server IP:PORT as an argument. Second argument can optionally be sending interval in seconds.
  • Use your computer normally until your computer crashes or whatever issue you are facing happens then restart.
  • Retrieve the logged data from the server's temporary memory by using one of the endpoints provided.

Endpoints:

  • GET: /export.json.gz - Export the data logged on the server's temporary memory as a gzip-compressed download (Recommended).
  • GET: /export.json - Export the data logged on the server's temporary memory as an uncompressed download.
  • GET: /export_to_internal_storage - Export the data logged on the server's temporary memory to the server's internal storage. (May not work depending on write priviliges)
  • POST (JSON): /sensor_logger - The client logger uses this to submit sensor data, you don't need to touch it.

Note: Using any of the export endpoints or closing the server app clears the logs from the server's temporary memory.

macOS and Windows support

The server side will work on both macOS and Windows thanks to Python, except for /export_to_internal_storage endpoint.
The client side will only work on Linux for now, until I find a solution that works on Windows.
macOS client support may never come though, as I don't own an Apple computer and probably never will.