[Support] Breadlysm - SpeedFlux

[breadlysm]

Overview: Support for the docker container SpeedFlux. SpeedFlux continuously runs SpeedTests and Latency tests on your network and stores the data in InfluxDB. 

App Name: SpeedFlux

App: https://github.com/breadlysm/speedtest-to-influxdb

GitHub Container: https://github.com/users/breadlysm/packages/container/package/speedflux

 

Feel free to leave feedback or suggest improvements. Here or a GitHub issue works. 

[mattekure]

I attempted to install this.  During the install the influxdb username and password fields could not be left blank, they required information to be added, which caused my authentication with influxdb to fail.  To complete the install I ended up deleting those two variables and recreating them, so they could be left blank.

[breadlysm]

1 hour ago, mattekure said:

I attempted to install this.  During the install the influxdb username and password fields could not be left blank, they required information to be added, which caused my authentication with influxdb to fail.  To complete the install I ended up deleting those two variables and recreating them, so they could be left blank.

That was an oversight on the template. The code already accepts blank user and password as you found. I did just fix the template though for future users so that those are no longer required. 

[mattekure]

So far I am loving it.  I got it integrated into my Ultimate UNRAID dashboard.  and it all seems to be running smoothly.  Thanks for making it.

[diymatt]

Oddly enough this is the only time anything has ever crashed my entire UnRaid machine.  

 

I installed it, watched the logs, saw I had a bad Influx connection on my end and then fixed it.  I saw data being written and then fired up my grafana instance and got that dashboard setup.  I noticed my network was offline so I checked unraid and it was 100% offline, the last thing I saw in the GUI which was locked up was 100% cpu usage across the board.  I thought it was a weird fluke so I went downstairs, rebooted the server and got back into grafana to play.  The crash happened again 5 mts later.  I changed the test time to 15 mts and it happened again in 15 mts.  

Thats way more unsafe shutdowns in 30 mts than I've had in years so I stopped the docker container and I've been running fine since.  

 

It may be coincidence, it may be my setup, but I wanted to at least comment about my experience in case it happened to anybody else.  

[breadlysm]

@diymatt Thanks for sharing and that's definitely weird. What you're describing sounds like some kind of infinite loop. 

 

Were you attempting to install before or after 3pm Eastern time? I'm wondering if a small change I made earlier introduced some type of unforgiving loop or if something screwy was happening before. Unfortunately I don't have a lot to look through to try and identify it what caused it but, it happening right when a test ran is a breadcrumb at least. If there's any other details you have, i'd be interested. 

 

Obviously I don't want it crashing entire machines.  It's probably worth me adding log files that persist so they aren't destroyed with a container. 

[diymatt]

@breadlysm I'm happy to report I re-tried it today am no longer experiencing system crashes.  W00t!  Very cool, thank you.

[Natcoso9955]

Awesome docker, i notice from your dashboard screenshot you had multiple servers, when i expand the extra params i see we can set a ookla ID, but im guessing only 1, is this correct?

EDIT: when i try multiple it doesnt use any, is this by design is my question.

Edited June 2 by Natcoso9955

[ehbush]

Been trying to make use of SpeedFlux for many days now! I would love to use it @breadlysm , but it doesn't seem to support InfluxDB when SSL is enabled. Are you able to add support for this?

[ehbush]

Hey @breadlysm any update(s) here?