Signal K Filter

[Lazzz]

I can prioritize the sentences that are doubled, tripled up as per the docs but how do you save it??

After I have been through & done them all & I restart Signal K they are all back there again!!

Is this normal?

[Lazzz]

Thanks for the replies!!

Is it because it's a stupid question or nobody knows??

[PaddyB]

Could you explain a bit more what you want to do?
What does 'prioritize'mean?
What extra 'sentences'? NMEA? Where do they come from?

[Lazzz]

Could you explain a bit more what you want to do?
What does 'prioritize'mean?
What extra 'sentences'? NMEA? Where do they come from?

From the Help docs:

What is Signal K Filter?[url=https://openplotter.readthedocs.io/en/latest/skfilter/SKfilter_app.html#what-is-signal-k-filter][/url]
The Signal K server wants to get all data it gets.
Why should we filter Signal K?

• Some devices send data which are everytime wrong (You can’t disable unwanted sentences on a nmea 2000 bus, or a defective water temp sensor in the log on nmea 0183, or …)
  
  
• You have a backup device (A second GPS or AIS)
  

To explain the problem we look at two GPS with a distance to each other.
If you zoom your nautical chart, your boat will constantly zigzag. Or one GPS has a bad reception.
The Signal K path will be identical but not the source.
The typical way would be to deside for every path which source you want to have. As long as you are able to make this decision everything is fine. (When you start converting nmea 2000 to nmea 0183 or the other way. Is there a chance to say only convert the Signal K path from a special source? No.
There are enough reasons to filter the data before it gets into the Signal K server.
There are two ways to filter.
In OpenPlotter SKfilter the page:

• Filter is the hard way. Only Signal K pathes witch match the criteria come to the server.
  
  
• Priority is the smooth way to filter. The criteria are the same, but if there is no communication within a time limit (timeout) that meets the criteria, any source for that Signal K path comes to the Signal K server. This is a kind of fallback functionality
  

[PaddyB]

What *exactly* are you trying to do and what is the problem?

[tkurki]

A solution to this is work in progress here https://github.com/SignalK/signalk-serve.../pull/1074

[Lazzz]

What *exactly* are you trying to do and what is the problem?

When I turn on my chart plotter, which turns on the CAN & sends heaps of info to SK & then it slows my Pi4 & eventually freezes!!

I will have 4 navigation.position.longitude for example or  4 navigation.speedOverGround plus a lot of other inputs from different sources doubled up all going to SK.

What I am trying to do is filter out/prioritize different inputs, that are doubled up, so as not to overload SK & so I can use my Pi4 when actually sailing!! 

As it is now, freezing up with too much conflicting info, it is not much good at all.

I can prioritize the doubled up inputs via the SK Filter as per the docs but there is no way to save the settings & after a re-boot they are all showing in the Diagnostic SK input window again - & there is no way of telling if the inputs are actually prioritized.

Having said, that I love OpenPlotter & appreciate all the hard work that goes into it - it runs 24/7 on my boat & I use it for monitoring various temps, sending me pics, letting me know if the bilge has water or there is movement (pir) etc etc.

There are no inputs from can in this pic - it is turned off except when boating - but there are still some doubled up.

   

---

A solution to this is work in progress here https://github.com/SignalK/signalk-serve.../pull/1074

Thank you very much

[e-sailing]

I can prioritize the sentences that are doubled, tripled up as per the docs but how do you save it??

After I have been through & done them all & I restart Signal K they are all back there again!!

Is this normal?

Filter and Priority are part of the file /home/pi/.signalk/red/flows_openplotter.json and are saved there.
After you delete/edit/add a Filter or Priority. This file will be changed. To activate the changes, you need to restart Signal K which restarts Node-RED.
(Please check if signalk-node-red is installed. (SignalK Dashboard Appstore->Installed))

How does it work?
Openplotter Signal K Filter only adds a flow "OpenPlotter Filter" to Node-RED and add some subflows with Java code. So it is managed under the hood of Signal K.
It makes use of the Signal K functions in Node-RED. And it is based on an example of Scott Bender (Signal K Slack forum. Thank you Scott!)


Yes it is normal. Last sentence of documentation warns.
"
Known issues
The Signal K server starts before Node-RED. A few seconds you will get data unfiltered.
"
So all your unwanted sentences (Filter/Priority) show up in the diagnostics but shouldn't update anymore!

[Lazzz]

I can prioritize the sentences that are doubled, tripled up as per the docs but how do you save it??

After I have been through & done them all & I restart Signal K they are all back there again!!

Is this normal?

Filter and Priority are part of the file /home/pi/.signalk/red/flows_openplotter.json and are saved there.
After you delete/edit/add a Filter or Priority. This file will be changed. To activate the changes, you need to restart Signal K which restarts Node-RED.
(Please check if signalk-node-red is installed. (SignalK Dashboard Appstore->Installed))

How does it work?
Openplotter Signal K Filter only adds a flow "OpenPlotter Filter" to Node-RED and add some subflows with Java code. So it is managed under the hood of Signal K.
It makes use of the  Signal K functions in Node-RED. And it is based on an example of Scott Bender (Signal K Slack forum. Thank you Scott!)


Yes it is normal. Last sentence of documentation warns.
"
Known issues
The Signal K server starts before Node-RED. A few seconds you will get data unfiltered.
"
So all your unwanted sentences (Filter/Priority) show up in the diagnostics but shouldn't update anymore!

Excellent!! Thank you so much

Yes, I have Node Red - with several flows running.