Chapter 4
Internal
This chapter gives you information that I wasn’t sure on where to put else.
Configuration Reference
adsf
Observation Data Layout
The original SatNOGS client opted for a single directory in /tmp/ that holds all (temporary) observation artifacts. While this work okay for most of the time, I found some drawbacks with this:
- Status of files is encoded in their file name: I especially dislike this for some of the data files, which are prefixed with
recording_while they are captured. - File name prefixes make code for handling more complex: Quite self-explanatory. You need to rename files, once they are finished.
With these observations in mind, satnogs-client-rs tries to be more simple and straight forward. First of all, all the data related to an observation (job) gets gathered in a separate jobs directory. At default, this is under /var/tmp/satnogs-client-rs/${JOB_ID}, where ${JOB_ID} is the unique number the network applied to your observation job.
Secondly, we signalise the status of a data directory by two files:
.processing_started: Processing the data via post-observation scripts (see Operation) has started (and may be still in process)..processing_finished: Running post-scripts has finished. Only once this file is present, the client will start uploading the artifacts to the network.
Example listing of the /var/tmp/satnogs-client-rs/ directory.