doc: Clarify the limits of dumping/backfilling via OpenMetrics

This is about native histograms (not yet supported) and staleness
markers (for which OpenMetrics support isn't even planned).

Signed-off-by: beorn7 <beorn@grafana.com>
This commit is contained in:
beorn7 2024-05-21 14:44:55 +02:00
parent 3119b8a055
commit 3127a4029e
3 changed files with 5 additions and 2 deletions

View file

@ -240,7 +240,7 @@ func main() {
dumpMaxTime := tsdbDumpCmd.Flag("max-time", "Maximum timestamp to dump.").Default(strconv.FormatInt(math.MaxInt64, 10)).Int64()
dumpMatch := tsdbDumpCmd.Flag("match", "Series selector. Can be specified multiple times.").Default("{__name__=~'(?s:.*)'}").Strings()
tsdbDumpOpenMetricsCmd := tsdbCmd.Command("dump-openmetrics", "[Experimental] Dump samples from a TSDB into OpenMetrics format. Native histograms are not dumped.")
tsdbDumpOpenMetricsCmd := tsdbCmd.Command("dump-openmetrics", "[Experimental] Dump samples from a TSDB into OpenMetrics text format, excluding native histograms and staleness markers, which are not representable in OpenMetrics.")
dumpOpenMetricsPath := tsdbDumpOpenMetricsCmd.Arg("db path", "Database path (default is "+defaultDBPath+").").Default(defaultDBPath).String()
dumpOpenMetricsSandboxDirRoot := tsdbDumpOpenMetricsCmd.Flag("sandbox-dir-root", "Root directory where a sandbox directory would be created in case WAL replay generates chunks. The sandbox directory is cleaned up at the end.").Default(defaultDBPath).String()
dumpOpenMetricsMinTime := tsdbDumpOpenMetricsCmd.Flag("min-time", "Minimum timestamp to dump.").Default(strconv.FormatInt(math.MinInt64, 10)).Int64()

View file

@ -585,7 +585,7 @@ Dump samples from a TSDB.
##### `promtool tsdb dump-openmetrics`
[Experimental] Dump samples from a TSDB into OpenMetrics format. Native histograms are not dumped.
[Experimental] Dump samples from a TSDB into OpenMetrics text format, excluding native histograms and staleness markers, which are not representable in OpenMetrics.

View file

@ -197,6 +197,9 @@ or time-series database to Prometheus. To do so, the user must first convert the
source data into [OpenMetrics](https://openmetrics.io/) format, which is the
input format for the backfilling as described below.
Note that native histograms and staleness markers are not supported by this
procedure, as they cannot be represented in the OpenMetrics format.
### Usage
Backfilling can be used via the Promtool command line. Promtool will write the blocks