--- title: Use Flux tasks with Kapacitor description: Use Kapacitor 1.6+ to run Flux tasks against InfluxDB and other data sources. Leverage the full library of Flux functionality to build powerful data processing and monitoring tasks in Kapacitor. url: https://docs.influxdata.com/kapacitor/v1/working/flux/ estimated_tokens: 6061 product: Kapacitor version: v1 --- # Use Flux tasks with Kapacitor Use **Kapacitor 1.6+** to run Flux tasks against InfluxDB and other data sources. Leverage the full library of Flux functionality to build powerful data processing and monitoring tasks in Kapacitor. ## Before you start Before you get started with Flux tasks, consider: - Kapacitor Flux tasks can not use Kapacitor topics or event handlers. You can only send alerts from within a Flux script using Flux notification endpoints. - Flux tasks are scheduled and executed by the Flux task engine built into Kapacitor 1.6+. This engine is separate from the Kapacitor TICKscript task engine. - Flux tasks are configured with the Flux `task` option inside of a Flux task script. This includes the task name and execution schedule. **Select which version of InfluxDB you’re using:** **InfluxDB 1.x:** 1. [Set up a Flux task database in InfluxDB](#set-up-a-flux-task-database-in-influxdb) 2. [Configure Kapacitor Flux tasks](#configure-kapacitor-flux-tasks) 3. [Create a Flux task](#create-a-flux-task) ## Set up a Flux task database in InfluxDB *(Optional but encouraged)* When Kapacitor executes a Flux task, it can store information about the task execution (run) in an InfluxDB database. To store this data, do the following: 1. Create a new database in InfluxDB to store Flux task run and log data in. ```sql CREATE DATABASE kapacitorfluxtasks ``` 2. To prevent large amounts of Kapacitor Flux task log data on disk, update the default `autogen` retention policy with a finite retention period or create a new retention policy (RP) with a finite retention period. **Update RP:** ```sql -- Syntax ALTER RETENTION POLICY ON DURATION -- Example ALTER RETENTION POLICY autogen ON kapacitorfluxtasks DURATION 3d ``` **Create new RP:** ```sql -- Syntax CREATE RETENTION POLICY on DURATION -- Example CREATE RETENTION POLICY threedays on kapacitorfluxtasks DURATION 3d ``` ## Configure Kapacitor Flux tasks Update or add the following settings under `[fluxtask]` in your `kapacitor.conf`: - **enabled**: `true` - **task-run-influxdb**: Name of the [InfluxDB configuration in your `kapacitor.conf`](/kapacitor/v1/administration/configuration/#influxdb) to use to store Flux task data. *To disable Flux task logging, set to `"none"`.* - **task-run-bucket**: InfluxDB database to store Flux task data and logs in. We recommend leaving this empty. By default, data is written to the `kapacitor_fluxtask_logs` database. To specify another database to write task log data to, use the `"db-name"` naming convention (including the retention policy `"db-name/rp"` is not supported). If the specified database does not already exist in InfluxDB, Kapacitor attempts to create the database. If authentication is turned on, permissions to `CREATE DATABASE` are required. For more information, see [Authentication and authorization in InfluxDB](/influxdb/v1/administration/authentication_and_authorization/). - Provide one of the following: - **task-run-org**: *Leave as an empty string (`""`)* - **task-run-orgid**: *Leave as an empty string (`""`)* - **task-run-measurement**: InfluxDB measurement to store task run and log data in. Default is `"runs"`. ##### Kapacitor Flux task configuration example ```toml # ... [fluxtask] enabled = true task-run-influxdb = "default" task-run-bucket = "kapacitor_fluxtask_logs" task-run-org = "" task-run-orgid = "" task-run-measurement = "runs" # ... ``` *For more information about Kapacitor `[fluxtask]` configuration options, see [Configure Kapacitor](/kapacitor/v1/administration/configuration/#flux-tasks).* ## Create a Flux task 1. Create a Flux task script. Include [the task option](/influxdb/v2/process-data/task-options/) in your script to configure the Kapacitor Flux task. *For more information about writing Flux tasks, see:* - [Get started with Flux tasks](/influxdb/v2/process-data/get-started/) - [Common data processing tasks](/influxdb/v2/process-data/common-tasks/) #### Provide InfluxDB connection credentials [`from()`](/flux/v0/stdlib/influxdata/influxdb/from/) and [`to()`](/flux/v0/stdlib/influxdata/influxdb/to/) functions require your InfluxDB **host** and **token**. - **host:** InfluxDB URL. - **token:** If **[InfluxDB authentication is enabled](/influxdb/v1/administration/authentication_and_authorization)**, use the `username:password` syntax. Otherwise, use an empty string (`""`) for your token. #### Bucket name syntax When querying or writing to InfluxDB 1.x with Flux, use the `database-name/retention-policy-name` pattern to specify your bucket. ##### example-task.flux ```js option task = { name: "example-task-name", every: 1h, offset: 10m } host = "http://localhost:8086" token = "" from(bucket: "example-db/example-rp", host: host, token: token) |> range(start: -task.every) |> filter(fn: (r) => r._measurement == "example-measurement") |> aggregateWindow(every: 10m, fn: mean) |> to(bucket: "example-db/example-rp-downsampled", host: host, token: token) ``` 2. Use the `kapacitor flux task create` command to add your Flux script as a Kapacitor Flux task. ```sh kapacitor flux task create --file /path/to/example-task.flux ``` *For more details about creating Kapacitor Flux tasks, see [Create a Kapacitor Flux task](/kapacitor/v1/working/flux/manage/create/).* **InfluxDB Cloud or 2.x:** #### Consider using InfluxDB tasks If you’re using **InfluxDB Cloud** or **InfluxDB OSS 2.x**, consider using [native InfluxDB tasks](/influxdb/cloud/process-data/) for data processing. 1. [Set up Kapacitor for InfluxDB Cloud or 2.x](#set-up-kapacitor-for-influxdb-cloud-or-2x) 2. [Configure Kapacitor Flux tasks for InfluxDB Cloud or 2.x](#configure-kapacitor-flux-tasks-for-influxdb-cloud-or-2x) 3. [Create a Flux task](#create-a-flux-task-v2) ## Set up Kapacitor for InfluxDB Cloud or 2.x Configure Kapacitor to connect to InfluxDB Cloud or InfluxDB OSS 2.x. For detailed instructions, see the following: - [Use Kapacitor with InfluxDB Cloud](/influxdb/cloud/tools/kapacitor/) - [Use Kapacitor with InfluxDB 2.x OSS](/influxdb/v2/tools/kapacitor/) ## Configure Kapacitor Flux tasks for InfluxDB Cloud or 2.x Update or add the following settings under `[fluxtask]` your `kapacitor.conf`: - **enabled**: `true` - **task-run-influxdb**: Name of the [InfluxDB configuration in your `kapacitor.conf`](/kapacitor/v1/administration/configuration/#influxdb) to use to store Flux task data. *To disable Flux task logging, set to `"none"`.* - **task-run-bucket**: InfluxDB bucket to store Flux task data and logs in. We recommend leaving this empty. By default, data is written to the `kapacitor_fluxtask_logs` bucket. To specify another bucket to write task log data to, use the [\_tasks system bucket](/influxdb/cloud/reference/internals/system-buckets/#_tasks-system-bucket) or [create a new bucket](/influxdb/cloud/admin/buckets/create-bucket/). If the specified bucket does not already exist in InfluxDB, Kapacitor attempts to create it with [`POST /api/v2/buckets`](/influxdb/v2/api/buckets/), in which case your API token must have permissions to create buckets in InfluxDB. For more information, see [Manage API tokens](/influxdb/v2/admin/tokens/). - Provide one of the following: - **task-run-org**: InfluxDB organization name. - **task-run-orgid**: InfluxDB organization ID. - **task-run-measurement**: InfluxDB measurement to store task run and log data in. Default is `"runs"`. ```toml # ... [fluxtask] enabled = true task-run-influxdb = "InfluxDB" task-run-bucket = "kapacitor_fluxtask_logs" task-run-org = "example-org" task-run-measurement = "runs" # ... ``` ## Create a Flux task 1. Create a Flux task script. Include [the task option](/influxdb/v2/process-data/task-options/) in your script to configure the Kapacitor Flux task. *For more information about writing Flux tasks, see:* - [Get started with Flux tasks](/influxdb/v2/process-data/get-started/) - [Common data processing tasks](/influxdb/v2/process-data/common-tasks/) #### Provide InfluxDB connection credentials `from()`\](/flux/v0/stdlib/influxdata/influxdb/from/) and [`to()`](/flux/v0/stdlib/influxdata/influxdb/to/) functions require your InfluxDB **host** and **token**. - **host:** InfluxDB URL. - **token:** If **[InfluxDB authentication is enabled](/influxdb/v1/administration/authentication_and_authorization)**, use the `username:password` syntax. Otherwise, use an empty string (`""`) for your token. ##### example-task.flux ```js option task = { name: "example-task-name", every: 1h, offset: 10m } host = "http://localhost:8086" token = "" from(bucket: "example-bucket", host: host, token: token) |> range(start: -task.every) |> filter(fn: (r) => r._measurement == "example-measurement") |> aggregateWindow(every: 10m, fn: mean) |> to(bucket: "example-bucket-downsampled", host: host, token: token) ``` 2. Use the `kapacitor flux task create` command to add your Flux script as a Kapacitor Flux task. ```sh kapacitor flux task create --file /path/to/example-task.flux ``` *For more details about creating Kapacitor Flux tasks, see [Create a Kapacitor Flux task](/kapacitor/v1/working/flux/manage/create/).* #### Related - [Use Kapacitor with InfluxDB Cloud](/influxdb/cloud/tools/kapacitor/) - [Use Kapacitor with InfluxDB OSS](/influxdb/v2/tools/kapacitor/) - [Create Kapacitor Flux tasks](/kapacitor/v1/working/flux/manage/create/) - [kapacitor flux](/kapacitor/v1/reference/cli/kapacitor/flux/) - [Get started with Flux tasks](/influxdb/cloud/process-data/get-started/) - [Common data processing tasks](/influxdb/cloud/process-data/common-tasks/) - [Task configuration options](/influxdb/cloud/process-data/task-options/)