If the Elementary on-run-end hook has stopped appearing in your logs after a recent deployment, there are a few common reasons for this behavior based on the knowledge sources: 1. Hooks Disabled via Variables: The most frequent cause is that the on-run-end hooks have been disabled, either intentionally or accidentally, in your

dbt_project.yml

or via command-line variables. Check for any of the following variables being set to

true

or to a value that disables them for your production target:

yaml    vars:      disable_run_results: true      disable_tests_results: true      disable_dbt_invocation_autoupload: true      disable_dbt_artifacts_autoupload: true

If any of these are set, the hooks will not run, and as a result, no data will be uploaded to the Elementary tables, leading to an empty dashboard. This is confirmed in several support threads where users saw empty reports after disabling these hooks, and re-enabling them resolved the issue [[Slack thread](https://slack.com/archives/C02CTC89LAX/p1697549788516149)]. 2. Elementary dbt Package Not Installed or Not Enabled: If the Elementary dbt package is not installed or is disabled in your

dbt_project.yml

, the hooks will not be registered and thus will not run. Ensure that your

packages.yml

includes Elementary and that your

dbt_project.yml

enables the package for the relevant target:

yaml    models:      elementary:        +enabled: "{{ target.name == 'prod' }}"

[[Slack thread](https://slack.com/archives/C02CTC89LAX/p1688568085309439)] 3. Profile or Target Misconfiguration: If your dbt run is using a different target or profile than expected, the hooks may not be configured to run, or they may be writing to a different schema/database. Double-check that your dbt and Elementary profiles are aligned and pointing to the correct schema/database [[Slack thread](https://slack.com/archives/C02CTC89LAX/p1708786310140179)]. 4. Order or Failure of Other Hooks: If you have multiple on-run-end hooks and one fails before the Elementary hook, dbt will not execute subsequent hooks. This can happen if, for example, you have a custom hook that errors out, preventing Elementary from running [[Slack thread](https://slack.com/archives/C02CTC89LAX/p1722936228529189)]. Recommended Actions: - Review your

dbt_project.yml

and deployment scripts for any of the disabling variables above. - Ensure the Elementary dbt package is installed and enabled for your production target. - Check that your dbt and Elementary profiles are correctly configured and aligned. - If you have custom hooks, verify that none are failing before the Elementary hook would run. If after these checks the hook still does not appear, try running a dbt command locally with increased log verbosity to see if any errors or warnings are being suppressed. Let me know if you need help reviewing your configuration!

If you have always had

disable_dbt_artifacts_autoupload: true

set and your dashboard was working before, then this variable alone should only disable the automatic upload of metadata models (like

dbt_models

,

dbt_tests

, etc.), but not the upload of run results or test results. The dashboard should still show test results and run results as long as the other related variables are not set to

true

[[on-run-end hooks documentation](https://docs.elementary-data.com/dbt/on-run-end_hooks#can-i-disable-the-on-run-end-hooks%3F)]. You do not need to set

disable_dbt_artifacts_autoupload

to

false

unless you want to re-enable the automatic upload of metadata models. If your dashboard relies on these metadata models being up-to-date, then yes, you should set it to

false

(or remove the variable) and see if the on-run-end hook resumes. If you only want to upload artifacts manually (for example, after deployments), you can keep it as

true

and run:

dbt run --select elementary.edr.dbt_artifacts

after each deployment to update the metadata [[on-run-end hooks documentation](https://docs.elementary-data.com/dbt/on-run-end_hooks)]. However, if nothing changed in your configuration and the dashboard suddenly stopped updating, it suggests something else may have changed in your environment or deployment process. Double-check that: - No other disabling variables (like

disable_run_results

,

disable_tests_results

, or

disable_dbt_invocation_autoupload

) were added or changed. - The Elementary dbt package is still installed and enabled. - The dbt run is using the expected target/profile. In summary: - You do not have to set

disable_dbt_artifacts_autoupload

to

false

unless you want to re-enable automatic metadata uploads. - If you want to keep it

true

, make sure to manually update artifacts as needed. - If your dashboard needs the metadata to be current and you want it to update automatically, set it to

false

. If the issue persists even though your configuration hasn't changed, there may be another cause outside of this variable. In that case, further investigation into your deployment or logs would be needed.

Your

+enabled

line:

yaml +enabled: "{{ target.name == 'elementary-prod' | as_bool }}"

may be problematic due to the placement of the

| as_bool

filter. In dbt, the

as_bool

filter should be applied to the entire expression, not just the string

'elementary-prod'

. The correct syntax should be:

yaml +enabled: "{{ (target.name == 'elementary-prod') | as_bool }}"

This ensures that the comparison is evaluated first, and then the result (

True

or

False

) is converted to a boolean. If you write it as you currently have, dbt will always convert the string

'elementary-prod'

to

True

, which is not what you want. This mistake could cause the Elementary models to be enabled in all environments, or not at all, depending on how dbt interprets the expression. This could explain why the on-run-end hook is not behaving as expected after your deployment. Recommendation: Update your

dbt_project.yml

to use the correct syntax:

yaml +enabled: "{{ (target.name == 'elementary-prod') | as_bool }}"

This approach is confirmed in multiple support threads and documentation as the correct way to conditionally enable Elementary models by target name [[FAQ](https://docs.elementary-data.com/oss/general/faq)], [[Slack thread](https://slack.com/archives/C02CTC89LAX/p1664996355098769)]. After making this change, re-run your dbt commands and check if the on-run-end hook appears in your logs and if the dashboard updates as expected.