Dear Grafana support team,
I am trying to pull info from my prometheus server. My metrics look the same all the time: labels which specifies, relevant information about some processes, and the value which is unix timestamp.
I use table view and i try to show every metric, so not just the actual, my current Grafana version is 11.3 .(How to extract label values from Prometheus metrics in Grafana | Grafana Labs)
Current status:
If i use: Verbose,Table format, Type instant then only the live metrics are visible in the table.
If i use Range then i can get back more metric labels but they are created via time stamps.
Thanks for the help, in advance.
Mate
That’s correct and exactly as Prometheus has documented it:
What is a problem here?
I would like to see all the previous metrics too, if i run the query on the prometheus server, and I change the time range every metrics are visible.
Could you provide example, pls?
I run the same query lets say: {processType=~“A”} from Prometheus in Graph view first in the time range 1s it only show activ metrics, If i change the range to 4week then I see all the previously active metrics as well.
In Grafana I only see the active ones, no matter how i change the timerange, the result is not changing(it is like only showing the latest live values of the metrics)
Sorry, it’s hard to understand. Provide reproducible example, printscreens,… Each tool/query ma have many options and you didn’t specify what have you condigured exactly.
It is restricted, company data i can not share the exact example, however i can share all the configs.
So in prometheus everything is on the base setup, I guess from Grafana side it is not really matter too, how the data arrives there and soon.
The actual metrics looks like this: processName{metricLabels=“(conatins timestampt too, to make it unique every time)”} metricValue==UnixTimeStamp
As the processName is constant when a new process start for the same process type, the prometheus metric values is overwritten and the previous lets say “live metric is stopped,closed”.
In grafana for me would be great to see all the previously running processes too
My grafana setup: Prometheus source, query options for the source are defult.
I use table view and for the individual query:i search for a specific label in the metrics I guess this one not really matters too about the topic(i dont use any specific time range setup or any prometheus related query feature, exactly run the same query as in the prometheus dashboard) and the Format is table, type is instant.
I already found the related docs for instant option and it says: Returns one data point per query and only the most recent point in the time range provided
So I would need something else, not the time range as it is not showing data individually, are there any other point that matters in this case? Please let me know and i will try to share data further more if it is needed
Actually it seems time range query for prometheus like [90d] is solving, the part of visibility now it shows every previous value too, but still time range is not working for the view as expected, if i change to now-1d then all the values are disappears.
