Hub went unresponsive again, because of InFluxDB

Yea I really want a trending solution.
Mine was ok until the Influx instance crashed.
Then I saw a storm of errors.

Hub was ok FWIW though...

I am not giving up influxDB. Love the visualizations. I guess I have been lucky and it hasn't caused me a single issue yet, and I log everything.

2 Likes

Remote reboot gone? Explain.....
I copied your Tasker profile for remote reboot.

See here:

And the root cause here:

Happy to report that my hub has been zippy again after disabling app logging.

I ran mine with log set to only display error and there was no errors in the log.

I initially couldn't get influx working so I had logging at Info. Forgot to change this so it was spamming the heck out of my logs.

yeah if you have too many devices enable it can do that. I just wish I know what's wrong with the app that's causing the hub lock up. I'm wondering if it's spamming the Hubitat with post request to the influxdb server.

My hub locked up Thursday morning as well, requiring a power cycle to restore it to working order. I am also running InfluxDB. Nothing in the logs.

That's disheartening. I was going to reach to you for advice on running on Windows instead of my Pi. Don't know if this is stable enough to bother anymore.

How many devices did you have pushing to influxDB?

I have about 35 devices reporting data updates to InfluxDB. It’s been working for many weeks without any issues. It seems strange that so many hubs locked up within such a short timeframe.

Are you running version 2 of the hub firmware? This happened to me when I didn't have influxdb logger installed also so now I am not sure what to think.

Yes, running v2 of Hubitat Firmware.

I currently have about 25 sensor only sending temperature readings. So far the hub has remained stable after disabling logging. My plan was to add devices (by capability) in groups and test for at least a day...but first I need to move this to a windows box because the the SD card issues using influx on a Pi might cause.
@ogiewon let us know if you find what locked you up. It may have just been f/w v2 related or a combo of influx on v2. I only had influx on v1 for a day so I have no point of comparison.

1 Like

My hub was locking up weekly and was told to remove influxdb. I plan on installing hublink and using the SmartThings hub to log to influxdb.

As a side note my hub has locked up since removing influxdb.

Interesting.. that's one way to get around the problem.

Ummm... so there's a stability issue between the influx app and HE so you're going to use ST because it's more stable???? That's Ironic....

2 Likes

Has anyone figured out a solution for this? I just had my first crash - not sure if there's an updated version of the influxdb app for hubitat.

The only method I'm aware of that doesn't seem to affect the hub is to tap into the eventsocket outlined below. Unfortunately it requires an external device of some kind to tap into the event stream and send out to a DB of your choosing. I have had this configured through nodeRed for months with no impact to my hub.

1 Like

Thank you, I saw that thread earlier and went with the simpler route instead. I'm actually used to nodeRED and have it running with HomeAssistant, so this should be an OK solution for now.

2 Likes