There is nothing Grafana can do if you specify a group … count(id) as "SLAMET" Could be useful regardless of the issue being fixed or not.I think you are misunderstanding how this works, if you have ave last 90 days and group by 10m and a small graph influxdb will return more data points than needed, and InfluxDB is really slow for large time ranges, (if you dont have rollups). Hello, I am looking to have my graph grouped by week, useful information here helps me to accomplish what we have here. but you can control how many bars you want in the template variable options.But "$interval" and "$summarize" are never passed to influxdb, are they? I can’t see field.keyword when I try to build Grafana chart with elasticsearch data source. week(created_on) as Week, On the right side of the query editor there is the option to switch to edit mode.This adds a 4day offset to the group by time.Do you know hw could I use this syntax as default ? This works as expected by adding $myvar to the "Group by time interval" field for my graphs.So, my question is how I could control the group by time interval through my templating variable, but still use time($interval) as a upper limit on the resolution. There is a syntax for getting what you want but i think you cant do it with the gui editor. GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.By clicking “Sign up for GitHub”, you agree to our.I've added my own templating variable with which I want to control the data point grouping resolution. FROM ticket The time period I'd display is relatively short, therefore retention is also quite short. INNER JOIN ( Otherwise one could argue that the group by time($interval) should always be in effect.I understand if there are no clean ways of adding this "feature". There doesn't seem to be any support for functions inside the group by field, e.g. Anyways, I am trying to make som time graph where I show top interface (or whatever) over time. I am looking to have my graph grouped by week, useful information here helps me to accomplish what we have here. Now at the time of the graph, I am looking to have the 2 values on the same group by WW, stacked.SELECT DISTINCT CONCAT(“WW”,week(T.created_on)) as metric,X.SLAEXP as value,Y.SLAMET as value1,Unix_timestamp(now()) as time_sec,__timeFilter(created_on) and status = 1 and datediff(updated_on, created_on) > 0 and projectId in(2) Both 10m and 30m result in sub-pixel resolution regardless of the series specific "group by" is set to "time(auto)" or "time($interval)".I would have liked to see the same performance and same graph for 10m and 30m if the "group by" was set to "time($interval)" when zoomed out far enough (e.g. Set fielddata=true on [user_name] in order to load fielddata in memory by uninverting the inverted index. Grafana/ClickHouse - GROUP BY with limit that summarizes “others ... Viewed 26 times 0. so, I am fairly new to Grafana and ClickHouse so please bear with me. For example the $interval is replaced by an actual time range that influxdb understands. Hi all, I have a strange behavior on my grafana panel. last 90 days). I have selected a Time range for the dashboard [ Jan 2016 to Feb end 2017 ] and my data points are only in Feb 2017.Successfully merging a pull request may close this issue.You signed in with another tab or window.http://play.grafana.org/dashboard/db/influxdb-templated-queries,http://play.grafana.org/dashboard/db/interval-template-variable. Still you can see the issue I'm trying to point out if you zoom out to last 90 days and toggle between 10m and 30m resolution. Their.I take it that Grafana does some pre-processing of the queries before passing them to influxdb.