The influx admin token is stored in the telegraf.conf file. No error message was displayed (InfluxDB was running as a service).
InfluxDB version: 2.8.0
Grafana-Enterprise version: 12.3.1
In Influx, data in the “Telegraf” bucket is queried and displayed normally. The same query in Grafana returns no data.
Data from the other bucket (ioBroker data) arrives in Grafana.
Immediately after restarting Influx, a message appears:
info Unauthorized {“log_id”: “10TtV0rl000”, “error”: “token required”}
The error occurs as soon as the Influx login attempt is made, i.e., when Influx is accessed in the browser.
Now also:
info Unauthorized {“log_id”: “10TvE5Zl000”, “error”: “session not found”}
Sorry, it seems these errors aren’t related to Telegraf. I stopped the Telegraf service and deleted the Telegraf bucket from Influx. The error still occurs. The connections between ioBroker, Influx, and Grafana are working correctly.
Nevertheless, I need to find the cause of the error.
When you check the same query in InfluxDB and Grafana, do you choose the same time period? I.e. Last 15 minutes?
Yes, I’ve already tested different time periods, including with various queries.
When I start the InfluxDB service, everything seems normal until I try to log in to InfluxDB via the browser. As soon as I try to log in, a message appears in the service window:
I also posted this problem in the Influx forum yesterday.
This message appears exactly once per hour, with the time shown in the image being one hour in the past.
I can process the data from ioBroker normally in Grafana, but no data from Telegraf is displayed in Grafana.
If you can get data from another InfluxDB’s bucket in Grafana, most likely this is not the access token issue…
Did you check Query inspector in Grafana? Are dates passed to InfluxDB from Grafana are correct?
Are InfluxDB and Grafana installed on the same computer? If not, do these computers have the same timezone configured?
Try hardcoding start and stop dates in InfluxDB UI query, and try the same query in Grafana.
I can only run new tests tomorrow. However, I can answer one question already.
InfluxDB and Grafana are running on the same PC and without Docker.
I’ve never used the query inspector in Grafana. I’ll do that tomorrow.
Is it correct that Grafana passes data (dates) to InfluxDB? I thought that only worked from InfluxDB to Grafana.
I’ll also set the date ranges (start and end) tomorrow.
Thank you very much in advance.
If you remove all filters from the query (keeping just bucket and time), still no data in Grafana?
So probably one of the filters in the query doesn’t work properly (may be, its value for some reason is not transferred correctly from Grafana to InfluxDB).
Try to find which one and adjust it.
As before, only the results are displayed without any filters.
However, I found something. In the view without all filters, there’s a reference to my still-active Raspberry Pi with the same system, but with current data. On this Windows 11 host, I have the old Raspberry Pi data. There’s no reference to the Raspberry Pi in the Telegraf.conf file.
I tried to delete the dashboards, but it’s not possible.
I was finally able to delete the dashboards.
I created a new dashboard, but it’s still pointing to the Raspberry Pi.
I don’t know where that comes from.
