New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Info: Signal-to-influxdb - successful setup with influxdb v2 and Raspian Bullseye 64-bit #61
Comments
I know it's been some time, but i'd be certainly interested in knowing how you acheived it. I cannot find a way to connect the plugin to influx v2. |
@pbfulmar would also like to know how you did it. I'd be willing to help get this upgraded to v2 as there are several reasons I want to go there and you just gave me another one (stable/cpu usage). My reason is that I want to do some local processing of datapoints as they come in and I can't do math across measurements without flux an v2 as I understand it. I've upgraded my Rasipi to bullseye and signalk seems to b performing fine. But, I have found the other influx add on to be insufficient as it really doesn't seem to support a localhost db (ie. informdb oss ver 2+). I'm a python guy and no expert on Node but it looks like:
Old client method -- influx.writePoints, influx.query Also, not sure whether tokens and organizations are needed for local/OSS mode. Most of the refactor seems to be in the index.js file in /src where the client is configured. If you would be willing to share, I could fork this and then post a pull-request for the author to update? Cheers! Lee PS - here is my source for seeing version support on the new javascript client drivers: |
@pbfulmar Hey please share how you got it working with Influxdb2. I am having difficulties getting it to work. |
Just for anyone else going through here: Found this article which I guess is what @pbfulmar did. In any case it worked for me, followed the article exactly (changed names and values obviously) and it worked great and didn't require any changes to the plugin itself. https://www.techetio.com/2021/11/29/influxdb-v2-using-the-v1-api-for-v1-dependent-applications/ TL;DR: execute the following on the influxdb machine
|
@amirlanesman - that was a big help. Worked like a champ! Journal dump:
Maybe I've inadvertenly added a scraper? Also a problem with derived data on one of my engines? The only thing that may look abnormal but would need someone more knowledgeable than me to verify:
Also, later, in journalctl, I see a few db errors but I think it may have been when I shutdown or very specific to the derived data plug-in. Will test these later:
On the influx journal, I see a few problems. Maybe a
Finally, I do hope the author rings in here. I think a better solution may be a slight refactor that supports the newer v2 client. I'm willing to try to fork it and put in a pull request but would like a little guidance on what parts of code i need to focus on. I assume its here: /src/skToInflux.js |
One other slight tweak. The startup of services and sequencing of signalk, influx, and grafana seem to be important on my Raspberry Pi 4B. I get db errors if signalk starts up before the influxdb. To resolve this, I added a unit definition in: /etc/systemd/system/signalk.service as follows:
this makes signalk wait until after the others are started and the errors go away. |
Hi guys, I just took a quick glimpse at the above mentioned article. Seems thats mostly what I did.
To configure the signalk-to-influxdb plugin of signalk, you need to supply Cheers |
Much thanks Peter. Yes, this was similar to the approach @amirlanesman found. I appreciate you pulling into a port and solving a mystery for us. :) Your full time cruiser status reminds me of Travis McGee. Here in the US, author John McDonald created the fictitious character "Travis McGee" who traveled abord the "Busted Flush" throughout Florida. He was a retired police detective who would solve a mystery in every port while "taking his retirement in installments." |
@leeprevost we could change the server setup script to include an option to specify/pick the services that SK depends on. So an option in the setup script that could add explicit dependency to |
I created a separate issue for real InfluxDb 2 support #63. |
Dear Teppo,
just wanted to let you know, that I have my whole SignalK setup now running on Raspi 64-Bit and influxdb v2.
With influxdb v1 the CPU usage recently was ramping up to more than 90%, now the system is only using 16% and that is absolutely stable since 6 days.
I‘m using the v1 API of influxdb v2, no modification to the SignalK influxdb plugin.
Let me know if you like to see the details of the influxdb setup.
Cheers
Peter
En route, SY Joy
The text was updated successfully, but these errors were encountered: