dbt run executes compiled sql model files against the current target database. dbt connects to the target database and runs the relevant SQL required to materialize all data models using the specified materialization strategies. Models are run in the order defined by the dependency graph generated during compilation. Intelligent multi-threading is used to minimize execution time without violating dependencies.

Deploying new models frequently involves destroying prior versions of these models. In these cases, dbt run minimizes the amount of time in which a model is unavailable by first building each model with a temporary name, then dropping the existing model, then renaming the model to its correct name. The drop and rename happen within a single database transaction for database adapters that support transactions.

Refresh incremental models

If you provide the --full-refresh argument to dbt run, dbt will treat incremental models as table models. This is useful when

  1. The schema of an incremental model changes and you need to recreate it
  2. You want to reprocess the entirety of the incremental model because of new logic in the model code
dbt run --full-refresh

In the dbt compilation context, this flag will be available as flags.FULL_REFRESH. Further, the is_incremental() macro will return false for all models in response when the --full-refresh flag is specified.

select * from all_events
-- if the table already exists and `--full-refresh` is
-- not set, then only add new records. otherwise, select
-- all records.
{% if is_incremental() %}
where collector_tstamp > (
select coalesce(max(max_tstamp), '0001-01-01') from {{ this }}
{% endif %}

Treat warnings as errors

In some cases, dbt will emit warnings rather than errors. These warnings are intended to alert you to potential issues with your dbt project, like the use of deprecated methods or configurations. To treat these warnings as errors (eg. in a CI environment) provide the --warn-error flag to your invocation of dbt.

$ dbt --warn-error run

Running specific models

dbt will also allow you select which specific models you'd like to materialize. This can be useful during special scenarios where you may prefer running a different set of models at various intervals. This can also be helpful when you may want to limit the tables materialized while you develop and test new models.

For more information, see the Model Selection Syntax Documentation.

Failing fast


Supply the -x or --fail-fast flag to dbt run to make dbt exit immediately if a single model fails to build. If other models are in-progress when the first model fails, then dbt will terminate the connections for these still-running models. In the example below, note that 4 models are selected to run, but a failure in the first model prevents other models from running.

$ dbt run --fail-fast --threads 1
Running with dbt=0.17.0
Found 4 models, 1 test, 1 snapshot, 2 analyses, 143 macros, 0 operations, 1 seed file, 0 sources
14:47:39 | Concurrency: 1 threads (target='dev')
14:47:39 |
14:47:39 | 1 of 4 START table model test_schema.model_1........... [RUN]
14:47:40 | 1 of 4 ERROR creating table model test_schema.model_1.. [ERROR in 0.06s]
14:47:40 | 2 of 4 START view model test_schema.model_2............ [RUN]
14:47:40 | CANCEL query model.debug.model_2....................... [CANCEL]
14:47:40 | 2 of 4 ERROR creating view model test_schema.model_2... [ERROR in 0.05s]
Database Error in model model_1 (models/model_1.sql)
division by zero
compiled SQL at target/run/debug/models/model_1.sql
Encountered an error:
FailFast Error in model model_1 (models/model_1.sql)
Failing early due to test failure or runtime error

Enable or Disable Colorized Logs


DBT can colorize the run logs output to your terminal to make the logs more readable. This is enabled by default, [can be overriden in your profiles.yml] (, and both the default and the profiles.yml can be overriden at the command line. Use --use-colors or --no-use-colors to enable or disable log colorizing from the command line. Example usage:

$ dbt --use-colors run
$ dbt --no-use-colors run

Deferring to previous run state

[β] Beta Feature

This is net-new functionality in v0.18.0, with iterative improvements to come. If you encounter unexpected behavior, please post in Slack or open an issue.

dbt run --models [...] --defer --state path/to/artifacts

When the --defer flag is provided, dbt will resolve ref calls differently depending on whether a node has been included in the model selection criteria of the current run. If a node is not included, references to it will resolve to the namespace specified by the state artifact.

This is especially useful when running a subset of models in a new test or development schema, as it obviates the need to create all of those models' parents.

Environment variables

Deferral and state can be set by environment variables as well as CLI flags:

  • --defer or DBT_DEFER_TO_STATE

If both the flag and env var are provided, the flag takes precedence.

Usage notes

  • In its current implementation, deferral requires state to be set explicitly, either by flag or env var. In a future release, we may set state to be the target-path by default.
  • The dbt run command (with no --models or --exclude specification) includes all nodes in its selection criteria, even seeds and sources. In this way, dbt run and dbt run --defer --state ... always yield identical results, with no deferred nodes.
  • Ephemeral models are never deferred, since they serve as "passthroughs" for other ref calls. Excluding an ephemeral model from the model selection criteria makes no difference to the output; including or excluding the non-ephemeral parent of an ephemeral model does.

Read more about state.


In my local development environment, I create all models in a schema named dbt_me. In production, the same models are created in a schema named prod. Both are located in a database called analytics.

I access the dbt-generated artifacts (namely manifest.json) from a production run, and copy them into a local directory called prod-run-artifacts.

from {{ ref('model_1') }}