I’m looking for some help to troubleshoot an issue I’m having when I have 4 or more Grafana graphs configured in a HABPanel widget which all show the last 7 days of data, I’ve only got about 4 days of persistance data captured so far.
I’m running the latest stable 2.5.0 via openHABian on a Pi 3+. I’m also using the Pride Theme if that’s relevant.
If I have 3 graphs configured the system runs without issue, if I configure a 4th the Pi will reboot within a few minutes, and continue to reboot once everything loads. There are no errors in the frontail log although I’ve only got the default level of logging setup.
What would be my first approach to troubleshoot this issue or is there a limitation with the number of graphs that can be configured within a single widget?
Confirm you are running InfluxDB, Grafana and openHAB all on the same Raspi3?
Will those 4 charts all be displayed at the same time?
When showing data for longer times, how many datapoints are being displayed (showing data for 7 days with a value for every X minutes)?
Do you log the load of your Raspi3?
@opus. Yes. I’m running openHab, Grafana and InfluxDB all locally on the same Raspi3. I’m also running the InfluxDB database locally, all from an SSD rather than SD.
All graphs display at the same time, currently persisting every minute, this is probably too often than needed. The Raspi load is one if the graphs in question so yes, I’m logging this too. Load rarely goes over 0.5. Added screenshot of the openHAB widget I have configured, without the much wanted 4th graph, Network System CPU Load (1min).
…and just as I wrote this reply it rebooted again with only the 3 graphs displaying.
One more thing, I’ve noticed the uptime has not changed via FireMotD so it’s like only openHAB is restarting rather than the Raspi itself.
For clarity, with no reported error, what makes you think openHAB is rebooting?
There are a number of reports of 2.5 unexpectedly apparently reloading items and/or things, hence the question.
Yes, logging halts, all web services, paper UI, basic UI, HABPanel, etc become unreachable and then all settings, items, rules etc reload.
I’ve now tried lowering all the graph periods to last 24 hours but the problem still occurs. openHAB restarts while HABPanel is attempting to load the graphs. So each time I access that specific widget if I have 4 or more (but sometimes also 3) openHAB restarts.
Log below of when I’ve edited the widget to add a 4th graph then accessed the HABPanel widget, within a minute it restarts or reloads everything.
2019-12-27 08:12:20.773 [INFO ] [panel.internal.HABPanelDashboardTile] - Stopped HABPanel
2019-12-27 08:12:21.353 [INFO ] [panel.internal.HABPanelDashboardTile] - Started HABPanel at /habpanel
2019-12-27 08:12:33.985 [vent.ItemStateChangedEvent] - networkpresence_responsetime01 changed from 133.0 to 125.0
2019-12-27 08:12:34.473 [vent.ItemStateChangedEvent] - networkpresence_lastseen01 changed from 2019-12-27T08:11:33.822+1000 to 2019-12-27T08:12:34.423+1000
2019-12-27 08:12:34.493 [vent.ItemStateChangedEvent] - d_presence_lastseen01 changed from 2019-12-27T08:11:33.833+1000 to 2019-12-27T08:12:34.438+1000
2019-12-27 08:13:03.767 [INFO ] [.core.internal.i18n.I18nProviderImpl] - Time zone set to 'xxxx'.
2019-12-27 08:13:03.809 [INFO ] [.core.internal.i18n.I18nProviderImpl] - Location set to '-xxxx,xxxx'.
2019-12-27 08:13:03.820 [INFO ] [.core.internal.i18n.I18nProviderImpl] - Locale set to 'en_AU'.
2019-12-27 08:13:12.373 [INFO ] [el.core.internal.ModelRepositoryImpl] - Loading model 'schedule.items'
2019-12-27 08:13:12.698 [INFO ] [el.core.internal.ModelRepositoryImpl] - Loading model 'presence.items'
2019-12-27 08:13:12.822 [INFO ] [el.core.internal.ModelRepositoryImpl] - Loading model 'house.items'
2019-12-27 08:13:12.897 [INFO ] [el.core.internal.ModelRepositoryImpl] - Loading model 'automation.items'