Nothing at all in either the openHAB log or the Grafana log? Weird. But you can use Grafana directly and get the graphs you want? Just not in BasicUI?
No nothing in the log⌠And yes, log in locally directly to Grafana, everything is working perfectly.
After changing Org.role to âViewerâ, it does seem like it´s working in BasicUI. At least I get an image now.
Unfortunatly, it doesnt seem to refresh/update, which makes it rather pointless 
What are you putting in the sitemap? In the 1st post of this thread there is some information on various ways to get it into the sitemap. The refresh should update the image.
This is from the sitemap.
Image refresh=60000 url="http://10.4.28.237:3000/render/d-solo/jwnxXmgRz/marc-temperatur?refresh=5s&orgId=1&panelId=2&from=1529258759230&to=1529269559230&width=1000&height=500&tz=UTC%2B02%3A00"
And this is how it looks. Screenshot is a few minutes old. Grafana picture hasn´t refreshed since 22.42 last night.
Itâs not going to change because youâve specified the absolute time for the time range:
You want this to me something like: â&from=now-48h&to=nowâ
1 Like
Ahh ofcouse⌠I thought the link was build from the time periode chosen like the same Grafana UI.
Now the image is updated and it seems to refresh allright. Thanks alot for your help, John!
I´ve got a new weird problem with GrafanaâŚ
I have created a dashboard for charting the humidity item from my Netamo indoor module.
This is the latest openhab event update from the item:
2018-06-20 22:59:06.753 [vent.ItemStateChangedEvent] - NetamoIndendRsHjemme_Humidity changed from 50 % to 51 %
But grafana dashboard is empty.
If I change the item, (just make a simple label change, save the item file), Grafana starts to read the data again and draw the chart, but just untill a few hours, then it ends up empty againâŚ
I know for sure there has been several updates picked up by OpenHab during the day. But still, Grafana is emptyâŚ
PS, forgot to mention, I have several other dashboards which works just fine. Only differences is, that the Netamo can be rather quite for 2-3 hours with the Humidity measuring,and therefore not sending any updated measuring.
This is Grafana settings. (Screendump)
You need to find out if the data was actually written to InfluxDB (implying a write problem) or if the data is written and is therefore a Grafana read problem from InfluxDB. What is your persistence strategy? Does it cover this item? You can turn on logging in InfluxDB (I think it is on by default) and see the data being written - logs to syslog. Do you see the updates being written for this sensor? InfluxDB has a command line interface - you can try the command in green in your screenshot. If it has no data, then Grafana wonât have any data points to display.
Hi JohnâŚ
My worries is, that it sometimes work, and sometimes it doesn´t.
It seems like Grafana stops charting and âdeleteâ all data for this item, after some hours without any updates. I know for sure grafana has written the chart within the last 24 hours, cause I saw it in the chart. How is it possible to dissapear all of a sudden then?
I will have a look at Influx to see if the data is actualy beeing written. But I would suspect it does. Never than less, it does not explain how this item can dissapear from the chart.
You should see the queries Grafana is using in the syslog as it queries InfluxDB. If you can replicate one of those and get data when your chart is empty, then I agree there is a Grafana problem. (And in that case youâd probably want to ask in the Grafana forum for a solution.) If you see the query going to InfluxDB and no data is returned, then there has to be an issue with the query being used or in the InfluxDB itself.
Hmm, I think I may have found something⌠But I have no idea what I´m looking at, to be honest. It seems like log is only wirtten for the specific item, when I view the Grafana dashboard. AnywayâŚ
First a screendump from the dashboard with the Netamo item right now, going 24 hours back:
According to this screendump, the first data querie is 00:45.
But acording to the syslog, the first entry is 23:10:06
Jun 20 23:10:06 openHABianPi influxd[750]: ts=2018-06-20T21:10:06.559405Z lvl=info msg="Executing query" log_id=08ff~SCl000 service=query query="SELECT mean(value) FROM openhab_db.autogen.NetamoIndendRsHjemme_Humidity WHERE time >= now() - 6h GROUP BY time(15s) fill(previous)"
Jun 20 23:10:06 openHABianPi influxd[750]: [httpd] 10.4.28.30, 10.4.28.30,10.4.28.237 - grafana [20/Jun/2018:23:10:06 +0200] "GET /query?db=openhab_db&epoch=ms&q=SELECT+mean%28%22value%22%29+FROM+%22lille_bad_fugt%22+WHERE+time+%3E%3D+now%28%29+-+6h+GROUP+BY+time%2815s%29+fill%28previous%29%3BSELECT+mean%28%22value%22%29+FROM+%22stort_bad_fugt%22+WHERE+time+%3E%3D+now%28%29+-+6h+GROUP+BY+time%2815s%29+fill%28previous%29%3BSELECT+mean%28%22value%22%29+FROM+%22Nilan_Input_RH%22+WHERE+time+%3E%3D+now%28%29+-+6h+GROUP+BY+time%2815s%29+fill%28previous%29%3BSELECT+mean%28%22value%22%29+FROM+%22NetamoIndendRsHjemme_Humidity%22+WHERE+time+%3E%3D+now%28%29+-+6h+GROUP+BY+time%2815s%29+fill%28previous%29 HTTP/1.1" 200 18704 "http://10.4.28.237:3000/d/_V-4oRggk/fugtighed?refresh=30s&orgId=1" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36" 4e653a92-74ce-11e8-b7bf-000000000000 495457
23:10 is where I messed around with the item, but grafana did not show the data in the dashboard.
I removed it from one dashboard and then created a new one. And it seems like Influx as well as Grafana does indeed have the dataâŚ
Now, how come Grafana first show the data in the new dashboard at 00:45?
Strange thing though, according to the screenshot above, it has been running stable ever since. 
I think I found the reason, well part of the reason.
If I set the dashboard to show the last 6 hours, and an item hasn´t updated in between, then Grafana will not show the item, even though it has been setup til âpreviousâ.
Thatâs why I asked about your persistence strategy. Are you only logging every change? You can also log periodically by time such as every hour. This might prevent the problem you are seeing.
If Grafana has no points in the time range you specify, it wonât have any previous value to display. It makes a query to the server with the given time span and tries to display whatever it gets back from the InfluxDB server. See the first post in the persistence section:
You´re absolutly correct John. The persistens file only log every change⌠Sorry I totally missed that part in your previous messageâŚ
Strategies {
everyMinute : "0 * * * * ?"
everyHour : "0 0 * * * ?"
everyDay : "0 0 0 * * ?"
every15min : "*/15 * * * * ?"
default = everyMinute
}
Items {
Temperatur* : strategy = restoreOnStartup, everyChange
Fugtighed*, gChart* : strategy = restoreOnStartup, everyChange
nilan_output_ExhaustSpeed, nilan_output_InletSpeed : strategy = restoreOnStartup, everyChange
}
Guess I could change it to:
Fugtighed*, gChart* : strategy = restoreOnStartup, everyChange, every15min
Then it should log every change and every 15 minute, correct?
maybe this should be?
"0 0/15 * * * ?"
I hope it works! 
1 Like
15 Minutes should be
"0 0/15 * ? * * *"
Edit:
Maybe i was too fast now and the syntax is different in rules, where i had this example from 
Hmm I have to trust you (both) in some way. I have no idea which is correct. My original was copied from another thread. And I havn´t figured out how to setup these times settings
According to http://www.quartz-scheduler.org/documentation/quartz-2.1.x/tutorials/tutorial-lesson-06
â0 0/15 * * * ?â
should fit.
I was shortly a bit confused because my expression has 7 parts and the on from @jswim788 has only 6 parts.
But it seems that the seventh part (stands for the year) is optional.
1 Like
Yes, my first one has a typo, it should have a 0/15 instead of the */15. And the second one is wrong - Iâll remove it from the post.
hi Guys
Ive installed Influx and Grafana. I can login to Grafana but I dont get a proxy option to connect to the Influx DB. I select âbrowserâ but then I get a network error.
Influx/Grafana are on the same host and are running fine
Any suggestions?