Starting from this week, we will try to maintain a regular SatNOGS client status meeting.
The meeting will take place every Thursday at 18:00UTC, on #satnogs channel @FreeNode IRC network (also accessible via riot.im). The agenda will include news, updates, discussions and decisions on client’s state and issues. The duration of the meeting will be kept as sort as possible.
- New way for generating waterfall system.
- It is under testing
- Needs some graphic changes
- It is much more efficient way (resources-wise) and much faster
- Needs changes in client code which are tracked by issue 292
- There was a decision to break compatibility with the old way of generating waterfalls
- Posting configuration of the station to network
- This task is tracked by issue 293
- example of data: client and gr-satnogs versions, lat/log etc… still need to determine what should be posted.
- Posting metadata for each observation
- This task is tracked by issue 294
- example of data: gains, sample rate, cpu/memory usage etc… still need to determine what should be included into these metadata
- Handle generated file and Incomplete observations(observations moved to incomplete due to an error while uploading)
- We decided to leave files into the complete directory in case of upload fail
- One exception when the observation doesn’t exist in network, then we should move it to incomplete.
- About generated files, crc fails and txt decoded data are not generated any more, so no need to handle them. This works in the latest gr-satnogs.
- Work on this can be tracked in issue 295
- Need of option for changing sample rate.
- This task needs changes on both client and gr-satnogs
- It’s not an easy task as it needs changes in some scripts in order to be parametric
- This issue is tracked by issue 296 for client and issue 115 for gr-satnogs
- There was an update on Ansible scripts to install latest tagged release of gr-satnogs and not master branch
- It makes gr-satnogs-package compile also tags in addition to nightly master builds
- You can change variables on Ansible to pick up the tagged version
- AFSK1k2 gr-satnogs script for decoding afsk1200 is under testing.
If I forgot something please ping me to include it.
PS As it was the first meeting after some time, it took a little longer than it should but the next ones will be sorter
Not many updates in this meeting as it was a busy week…
- AFSK1k2 was successful in Apomachon in production, check observations 48546 of NO-84 and 47951 of ECAMSAT.
- There is a pull request for supporting this gnuradio script on client, so it will be soon landed
- Script still needs some performance fixes but it works fine for pushing it to production.
- VOLK profile for better performance of gnuradio scripts.
- There is a thought to include the profile for rpi
- Maybe it doesn’t make such a difference for arm architecture
- Need some experiments to see the difference with it and without it
- CW gnuradio script
- There is an implementation to be tested
- Need an extra parameter for wpm
- There will be changes in network and client to support this extra parameter.
- Station metadata (not to be confused with metadata which will be sent for each observation)
- As changes come and in order to have a metric about fragmentation of the stations software we need to prioritize station metadata, especially those about software versions.
- I’m going to open a thread in the next days about it. We will need to write down which metadata we need to post back to network about the station
No updates in this meeting there were two discussions, the one was about the recent issue with usrp sdr which throws some errors and the second one about the metadata that should be posted from client to network. For the first one we think that is new firmware tha causes the issue, still investigating.
The regular IRC meeting is back: Regular IRC meeting for SatNOGS