Geomap shows marker only after entering and leaving edit mode - without actually changing something

info90f4 · May 9, 2025, 11:34am

Hi,

I observed several times the following behavior of the Geomap panel.
When I open the related dashboard the panel shows the map but not the expected position marker.
When I open the panel for edit, the preview in edit mode shows immediately the marker at the correct position. I close the edit view without changing anything and now the marker is correctly displayed in the normal dashboard view.

Some more information:

The dashboard has two Geomap panels. It seems that it is always the 2nd panel (from the top) that has the missing marker
Position data is queried from data in an InfluxDB database
Grafana version 11.5.2, running on Ubuntu (in a docker container)

BR Christian

yosiasz · May 9, 2025, 1:09pm

hello @info90f4

do you see the same issue if you use Orchestra City plugin?

info90f4 · May 9, 2025, 1:37pm

wasn’t aware of that plugin, so I can not answer your question.
Do you recommend to try it? Do you think there is a chance that it will fix the issue?

yosiasz · May 9, 2025, 1:43pm

I highly recommend it. will it fix it? Only one way to find out install it and try it out

info90f4 · May 9, 2025, 2:04pm

Ok, perfect!
I will try it as soon as I find time. When I briefly looked at the description it seems to provide also some other features!

Thank you so far!

info90f4 · June 26, 2025, 4:23pm

@yosiasz , it took a while, but I followed your suggestion and gave the Orchestra City plugin a try.
Indeed it seems that it does not show the described behavior.
Therefore it has another issue

If I wait it bit, this error is shown in the map:

If I wait again or reload the dashboard the error is gone and the map is shown correctly.
Maybe you can’t have everything?

yosiasz · June 26, 2025, 5:08pm

what does your datasource look like?

info90f4 · June 26, 2025, 5:21pm

It’s InfluxDB V2.7 (if that is the answer you are looking for?).

yosiasz · June 26, 2025, 5:25pm

sorry, I meant to say what does the csv data look like from influxdb in order to emulate your scenario and test things out.

info90f4 · June 26, 2025, 5:45pm

Ah, OK, here is the csv-data of the last 30 days.

Äh, it is not, the forums software does not allow to upload files. How can I do that?

Maybe already some background about the nature of the data. The sensor send it’s normal data regularly (e.g. once per hour). The GPS position is only sent when the sensor is moved. That means it can be, that I have to look far back in the data, to find the last sent GPS position.
Currently it happened 2 days ago.

yosiasz · June 26, 2025, 5:46pm

location,lon,lat
aqui,32,21

info90f4 · June 26, 2025, 7:55pm

Not completely sure what you mean? Post the csv directly here?

+ACM-group,false,false,true,true,false,false,true,true,true,true
+ACM-datatype,string,long,dateTime:RFC3339,dateTime:RFC3339,dateTime:RFC3339,double,string,string,string,string
+ACM-default,mean,,,,,,,,,
,result,table,+AF8-start,+AF8-stop,+AF8-time,+AF8-value,DeviceSerialNumber,+AF8-field,+AF8-measurement,host
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:10:05Z,52.112266,71556111,latitude,vega+AF8-flooding,owdata.de
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:20:05Z,52.112266,71556111,latitude,vega+AF8-flooding,owdata.de
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T18:00:05Z,52.112266,71556111,latitude,vega+AF8-flooding,owdata.de
,,1,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:10:05Z,9.356095,71556111,longitude,vega+AF8-flooding,owdata.de
,,1,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:20:05Z,9.356095,71556111,longitude,vega+AF8-flooding,owdata.de
,,1,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T18:00:05Z,9.356095,71556111,longitude,vega+AF8-flooding,owdata.de

yosiasz · June 27, 2025, 2:08am

Could you please now share your flux query and what the data looks like when queried for example

from(bucket: "überflutet")
  |> range(start: v.timeRangeStart, stop: v.timeRangeStop)
  |> filter(fn: (r) => r["_measurement"] == "vega+AF8-flooding")
  |> filter(fn: (r) => r["_field"] == "device" or r["_field"] == "latitude" or r["_field"] == "longitude")
  |> keep(columns: ["_time", "_value", "_field"])
  |> pivot(rowKey:["_time"], columnKey: ["_field"], valueColumn: "_value")

info90f4 · June 27, 2025, 8:04am

I’m using actually InfluxQL for querying the database.

Here are both queries.

Map 1

SELECT Last("latitude") AS "latitude", Last("longitude") AS "longitude" FROM "vega_flooding" WHERE ("DeviceSerialNumber"::tag = '71556111')

grafik
If I look up the data in the InfluxDB UI I find this as the last entry.

table,_time,_value,DeviceSerialNumber,_field,_measurement,host
0,2025-06-24T18:00:05Z,52.112266,71556111,latitude,vega_flooding,owiot.de
1,2025-06-24T18:00:05Z,9.356095,71556111,longitude,vega_flooding,owiot.de

Map 2

SELECT Last("GNSSLatitude") AS "latitude", Last("GNSSLongitude") AS "longitude" FROM "coop_garden" WHERE ("devEUI"::tag = '7074140000ab4546')

grafik

InfluxDB UI

table,_time,_value,_field,_measurement,devEUI,deviceName,host,location,municipality,sensortype,usecase
0,2025-05-15T21:02:10Z,52.11185073852539,GNSSLatitude,coop_garden,7074140000ab4546,VEGA_Air41-2-OW,owiot.de,OW,Hameln,Pegelsensor_Straße,Ueberschwemmung
1,2025-05-15T21:02:10Z,9.355592727661133,GNSSLongitude,coop_garden,7074140000ab4546,VEGA_Air41-2-OW,owiot.de,OW,Hameln,Pegelsensor_Straße,Ueberschwemmung

Both use different InfluxDB databases (“buckets”).

What puzzles me a bit is the time stamp in Grafana in both cases

It’s hard to examine the query result in Grafana,when there is no marker shown. As soon as I open the visualization in edit mode to look for the data, the marker appears and the data is as shown above!

info90f4 · June 27, 2025, 8:18am

As an Addition I did a small statistical test.
I setup a dashboard with 2 “Ochestra city map”-visualizations and 3 standard Grafana “Geomap” visualizations.
I reloaded the dashboard 20 times. Panels are numbered from top left to bottom right.

Ochestra city map

Markers on panel 1, 2 : 100%

I think we can call that reliable.

Geomap

Markers on panel 1 : 20%
Markers on panel 2 : 20%
Markers on panel 3 : 5%
Markers on panel 1, 2 : 5%
Markers on panel 1, 3 : 25%
Markers on panel 2, 3 : 25%
Markers on panel 1, 2, 3 : 0%

I think we can call that random.

info90f4 · June 27, 2025, 8:39am

I think I found out, what causes the error message for the “Ochestra city map”

It happens only when I hover with the mouse over the position marker.
It disappears (map shown correctly again) when I press the (data) refresh button in Grafana or wait till the automatic refresh.
In the settings there was a value defined for Position layer/Properties (Select properties to be displayed) that I did not query. When I remove that entry the error message does no longer happen!

info90f4 · June 27, 2025, 9:07am

@yosiasz , I think you pointing to the Orchestra City plugin helped a lot!
Thanks again.
I added the correct value to the query and have now a nice pop-up when hovering over the marker.

The only thing remaining is the wrong time-stamp. Do you have an idea what might cause that problem?

yosiasz · June 27, 2025, 1:55pm

info90f4:

result,table,+AF8-start,+AF8-stop,+AF8-time,+AF8-value,DeviceSerialNumber,+AF8-field,+AF8-measurement,host
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:10:05Z,52.112266,71556111,latitude,vega+AF8-flooding,owdata.de
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T16:20:05Z,52.112266,71556111,latitude,vega+AF8-flooding,owdata.de
,,0,2025+AC0-05+AC0-27T17:30:47.319935735Z,2025+AC0-06+AC0-26T17:30:47.319935735Z,2025+AC0-06+AC0-24T18:00:05Z,52.112266,715561

I see some odd looking date time stamps here wonder if that is the issue

info90f4 · June 27, 2025, 3:12pm

I think that is only a question of character formats.
I exported a csv-file from the InfluxDB UI, opened it in LibreOffice Calc, removed unneeded entries, saved it and opened it again in a text editor.
At that time underscores (“_”) and hyphens (“-”) where coded as +AF8- and +AC0-.

If I open the csv-file after download directly with a text editor it looks for example like this:

,result,table,_start,_stop,_time,_value,_field,_measurement,devEUI,deviceName,host,location,municipality,sensortype,usecase
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T05:02:00Z,4.573709487915039,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T06:56:00Z,3.507211446762085,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T09:02:00Z,-4.471893310546875,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T11:00:00Z,5.042413711547852,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T12:54:00Z,58.2926025390625,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:00:00Z,58.29380798339844,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:06:00Z,58.421302795410156,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:22:00Z,58.32550811767578,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:28:00Z,58.23210525512695,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:36:00Z,58.4452018737793,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T13:48:00Z,58.37150192260742,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T14:02:00Z,58.49240493774414,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T14:10:00Z,4.306714057922363,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung
,,0,2025-06-27T03:10:04.154977119Z,2025-06-27T15:10:04.154977119Z,2025-06-27T14:56:00Z,3.976702928543091,ScaledValue,coop_garden,7074140000ab4546,VEGA_Air41-2-OW-Lora,owiot.de,OW,Bad Pyrmont,Pegelsensor_Straße,Ueberschwemmung

Looks OK to me!

info90f4 · July 10, 2025, 1:00pm

I think I found out why I have the odd time stamp (where “01.01.1970, 01.00:00” is basically InfluxDB’s way to say: " I have no proper timestamp").
Here is my query in InfluxQL:

SELECT 
   last("GNSSLatitude") AS "latitude",
   last("GNSSLongitude") AS "longitude",
   last("ScaledValue") AS "Pegel" 
FROM "coop_garden" 
WHERE ("devEUI"::tag = '70xxxxxxxx46')

As far as I understood this query does not give a proper time stamp since “GNSSLatitude” and “GNSSLongitude” are only rarely written to data sets (when the sensor that provides the data, is moved). But “ScaledVale” ist written to each data set.
Hence

   last("GNSSLatitude") AS "latitude",
   last("GNSSLongitude") AS "longitude",

may provide data that is written to the database 3 month ago, whereas

  last("ScaledValue") AS "Pegel"

might give data that is written 10 minutes ago.

I was not able to come up with a query that returns:

last latitude and longitude
last “ScaledValue”-value
proper timestamp, taken from the last “ScaledValue”-value

As far as I understand the “Orchestra City Map” plugin, I can not split data (longitude/ latitude, value to display in popup) for the popup (in the “Position Layer”) to different queries (“A”, “B”, “C”).

My workaround is now to switch of the time stamp in the popup (remove “Pop up Time” entry). Not perfect, but OK.

Topic		Replies	Views
Geomap panel not working for me InfluxDB geomap-panel , influxdb	16	5726	January 16, 2024
GeoMap, new feature requests, influx work-arounds Geomap Panel geomap-panel , influxdb	17	2087	June 27, 2022
InfluxDB query results are not plotted correctly on Geomap InfluxDB geomap-panel	14	619	May 19, 2023
Grafana Geomap Route Geomap Panel influxdb	37	5022	August 28, 2024
Orchestra Cities dont plot data on maps Grafana plugins , geomap-panel	2	352	January 31, 2024

Geomap shows marker only after entering and leaving edit mode - without actually changing something

Related topics