The datastore_timeseries from me-box-archive

datastore_timeseries's Introduction

#Timeseries Datastore

This is a timeseries datastore for the databox

the datastore exposes a http based api on port 8080 withe following end points

Method: POST
URL: /api/reading
Parameters: Raw JSON body containing elements as follows {sensor_id: <sensor ID>, vendor_id: <vendor ID>, value:  <sensor value>}
Notes: The vendor_id and sensor_id must be valid and related entries in the databox directory in order for the reading to be accepted

Method: POST
URL: /api/reading/latest
Parameters: Raw JSON body containing elements as follows {sensor_id: <sensor ID>
Notes: will return the latest reading based on the databox wide sensor id

Method: POST
URL: /api/reading/latest
Parameters: Raw JSON body containing elements as follows {sensor_id: <sensor ID>}
Notes: will return the latest reading based on the databox wide sensor_id

Method: POST
URL: /api/reading/range
Parameters: Raw JSON body containing elements as follows {sensor_id: <sensor ID>, start: <start timestamp>, end: <end timestamp>}
Notes: will return the range of readings between start and end based on the databox wide sensor_id

TODO:

Implement checking of sensor id presence within databox directory for posting sensor readings
Implement checking of valid macaroon data to allow for access to latest and range data
Implement aggregated range reading so for example one could specify the reslution at which to return data (assuming complicance with SLA). This needs some thinking as not all timeseries data lends itself to all types of aggregation, i.e. state on and off cannot be averaged over time, likely that sensor meta data will be needed to added to the directory which describes the types of aggregation that are supported by the sensors data stream.

datastore_timeseries's People

Contributors

Watchers

datastore_timeseries's Issues

HTTPS Api

API moving to HTTPS with cert signed by the CM

Audit log

Audit log, to recover information about accesses made to/from the store.

Log all interaction into an audit DB.

Add the /audit end points

Suggestions:

1 Make it a time series like endpoint with /latest and /since endpoints  
2 Filter by sensor ID
3 Filter by sensor type 
4 Filter by reads and writes.