metrics.timeseries

metrics.timeseries data queries aggregates metrics statistics over a date range. This can retrieve multiple series each with a different metric and function (avg, sum, min, max, count). A series can also be aggregated or filtered by any combination of metric dimensions.



type:metrics.timeseries
period:hour
range:"-24 hours to now"
series.avg:(
  label:Average
  metric:cerb.tickets.open
  function:avg
)
series.min:(
  label:Min
  metric:cerb.tickets.open
  function:min
)
series.avg:(
  label:Max
  metric:cerb.tickets.open
  function:max
)
timezone:Europe/Berlin
format:timeseries

period:
range:
series.*:
- series.*.query:
timeout:
timezone:
format:

period:

The period is the metric’s level of detail:

`minute`	5-minute statistics (up to the past 24 hours)
`hour`	hourly statistics (up to the past 2 weeks)
`day`	daily statistics (no limit)
`week`	weekly statistics starting on Monday (no limit)
`week-sun`	weekly statistics starting on Sunday (no limit)
`month`	monthly statistics (no limit)
`year`	yearly statistics (no limit)

range:

The date range to show statistics for, grouped by the period.

this week

this month

-1 year

Jan 1 2025 to Dec 31 2025

series.*:

Each series.* series should provide:

Key	Description
`metric:`	The metric name
`by:`	A comma-separated list of dimension keys to group by
`function:`	`sum`, `min`, `max`, `average`, `samples`, `distinct`
`label:`	A human-readable label for this series
`missing:`	How to handle missing sample periods: `null` (default), `zero` (set to 0), or `carry` (carry the last sample)
`query:`	An optional query using dimension keys to filter samples.

series.*.query:

Dimension filters can use negation. For instance, query:(worker_id:![1,2,3])

Record-based dimension can use deep search filers.

timeout:

The query will be aborted after this many milliseconds have elapsed with no response.

The default is 20000 milliseconds (20 seconds).

timezone:

Statistics are stored in GMT/UTC. The timezone: location shifts timestamps when grouping metrics by period.

For example:

America/Los_Angeles

Europe/Berlin

Asia/Tokyo

If omitted, this defaults to the timezone of the current worker or the server.

format:

The metric results can be returned in various formats:

Format
`timeseries`	Returns series-based data suitable for a time-series chart (with the ‘x’ axis values as timestamps)

Cerb

Metrics Timeseries