docs-v2/content/shared/influxdb-v2/reference/cli/influx/bucket-schema/update.md

6.0 KiB

The influx bucket-schema update command updates the schema of an InfluxDB bucket that has the explicit schema-type.

bucket-schema update requires a bucket with at least one defined schema.

Usage

influx bucket-schema update [flags]
Supported operations
  • Adding new columns to a schema
Unsupported operations
  • Modify existing columns in a schema
  • Delete columns from a schema

Flags

Flag Description Input type {{< cli/mapped >}}
-c --active-config CLI configuration to use for command string
-n --bucket ({{< req >}}) Bucket name (mutually exclusive with --bucket-id) string
-i --bucket-id ({{< req >}}) Bucket ID (mutually exclusive with --bucket) string
--columns-file ({{< req >}}) Path to column definitions file. For more information, see Create a columns file. string
--columns-format Columns file format (csv, ndjson, json, default: auto). For more information, see Update a schema with columns format string
--configs-path Path to influx CLI configurations (default ~/.influxdbv2/configs) string INFLUX_CONFIGS_PATH
-x --extended-output Print column information for each measurement schema (default: false)
-h --help Help for the create command
--hide-headers Hide table headers (default false) INFLUX_HIDE_HEADERS
--host HTTP address of InfluxDB (default http://localhost:8086) string INFLUX_HOST
--json Output data as JSON (default false) INFLUX_OUTPUT_JSON
-n --name ({{< req >}}) Measurement name string
-o --org Organization name (mutually exclusive with --org-id) string INFLUX_ORG
--org-id Organization ID (mutually exclusive with --org) string INFLUX_ORG_ID
--skip-verify Skip TLS certificate verification INFLUX_SKIP_VERIFY
-t --token API token string INFLUX_TOKEN

Examples

{{< cli/influx-creds-note >}}

Update a schema using the influx CLI

influx bucket-schema update \
  --bucket example-bucket \
  --name temperature \
  --columns-file columns.csv

Update a schema and print column information

influx bucket-schema update \
  --bucket example-bucket \
  --name cpu \
  --columns-file columns.csv \
  -extended-output

Update a schema with columns format

By default, InfluxDB attempts to detect the columns file format. If your file's extension doesn't match the format, set the format with the columns-format flag.

influx bucket-schema update \
  --bucket example-bucket \
  --name cpu \
  --columns-file columns.json \
  --columns-format ndjson
influx bucket-schema update \
  --bucket example-bucket \
  --name cpu \
  --columns-file columns.txt \
  --columns-format csv