SatNOGS station monitor v0.4.x

SatNOGS Software
21

Hi @adamk,

  1. Yes, that is correct. I’m aware of this but haven’t tracked it with an issue yet. I now opened one here. We currently don’t support monitoring observations already in progress when starting the monitor. However it should be possible to implement this.

  2. The rotator position is polled every second. There is an open issue to increase the interval or at least make it configurable. maybe you can try increasing the interval as described in the issue and report back with your findings, or give further input about the rotator you’re using?

22

I released version 0.4.2. It should fix the issue regarding the rotator position update. The polling interval is now configurable in the config file or as command line argument. The default interval is now increased from 1 second to 5 seconds. See the release page for a full changelog.

You can update the installation on your station using:

wget https://github.com/wose/satnogs-monitor/releases/download/0.4.2/satnogs-monitor_0.4.2_armhf.deb
sudo dpkg -i satnogs-monitor_0.4.2_armhf.deb

Any feedback regarding the rotator issue is highly appreciated.

4 Likes
23

I greatly appreciate your quick action, wose! I have tested the 0.4.2. I can’t say the problem is solved - it behaves similar. I’ve connected using rotctl -m 2 -r 127.0.0.1:4533 and tried polling the position. Outside the observation the reply is always present and correct. During observation that’s another story:
Rotator command: p
get_pos: error = IO error

Rotator command: p
get_pos: error = Communication timed out

Rotator command: p
Azimuth: 53.000000
Elevation: 7.000000

Rotator command: p
get_pos: error = IO error

Rotator command: p
Azimuth: 53.000000
Elevation: 7.000000

Rotator command: p
get_pos: error = IO error

Rotator command: p
Azimuth: 49.000000
Elevation: 6.000000

Rotator command: p
Azimuth: 49.000000
Elevation: 6.000000

Rotator command: p
get_pos: error = Communication timed out

Anyway, I’d expect the satnogs-monitor to catch up at least some time. But the position is frozen either to the last successful reading, or to -1, -1. Looks to me like it gives up on some of the errors and does not retry?
I have tried changing satnogs_rot_threshold from 1° to 3°, so there would be wider time slot to poll the rotator position. That seems not to help. I got less frequent errors when manually polling position via rotctl.

24

Hi,

thx for testing. Could you try increasing the rotctld_interval to something like 10 or 15 to see if there is any change? Like:

satnogs-monitor --rotctld-interval 10 ... # rest of your arguments
25

I have finally discovered the log ‘l’ command. :slightly_smiling_face: in that window it says Lost connection to rotctld. Increasing interval to 10 did not help. It works the first 2 or 3 updates, than freezes…

26

I created a new issue here. I’ll need to check what’s rotctld is actually doing to understand the failure while polling the rotator position. Maybe this can be fixed on rotctld side.

1 Like
27

Hi adamk,
Just noticing that you say your connected using rotctl -m 2 -r 127.0.0.1:4533.
Should that be rotctld -m 2 -r 127.0.0.1:4533.
I’m no expert, Just hope it helps.
Steve

28

Hi Steve,
the rotctld deamon is running on the background already. I am using rotctl to talk to it via TCP. I believe that’s right. It is working unless more clients are trying to talk to 1 SPID rotator simultaneously. I think we really need to take a look at the rotctld code.
Thanks for your suggestion anyway.
AdamK