Hello. I ran this command (airflow) π airflow/sr...
# ask-elementary-ai
g
Hello. I ran this command (airflow) π airflow/src/plat/dbt 287-data-quality-tests-cli-tool-and-mcp ✗ ❯ edr send-report \ --profile-target prod \ --env prod \ --profiles-dir . \ --project-dir . \ --gcs-bucket-name elementary-reports-helcim \ --update-bucket-website true \ -d 1 Right after running (airflow) π airflow/src/plat/dbt 287-data-quality-tests-cli-tool-and-mcp ✗ ❯ dbt test --selector eq_dq_models --target prod --threads 8 And that command output 195905 2 of 2 START hook: elementary.on-run-end.1 ..................................... [RUN] 195905 2 of 2 OK hook: elementary.on-run-end.1 ........................................ [OK in 992.68s] If I query the elementary tables, I will find the data for the executed tests select id, test_execution_id, test_unique_id, model_unique_id, invocation_id, detected_at, database_name, schema_name, table_name, column_name, test_name, test_type, test_sub_type, severity, status, failures from elementary.elementary_test_results where table_name like 'dq_%' order by detected_at desc limit 100; But the report will not show those executions. What am I doing wrong? These tests live in a new dataset I created, before configuring Elementary for the first time
k
A message from kapa.ai
Based on your description, it sounds like you are running your dbt tests and seeing the results written to the
elementary.elementary_test_results
table, but the Elementary report generated by
edr send-report
is not displaying those test executions. This issue is commonly related to a mismatch between the database/schema where the test results are written and the configuration of the
elementary
profile used by the CLI to generate the report. Here are the most relevant points from the knowledge sources: 1. Profile and Schema Alignment: The
elementary
profile in your
profiles.yml
must point to the same database and schema where the
elementary
dbt package writes its results. If you have created a new dataset (schema) for your tests, ensure that both your dbt runs and the Elementary CLI are referencing this same location. If the CLI is pointed at a different schema, it will not find the new test results, even though they exist in the warehouse. Double-check that the
schema
and
database
fields in the
elementary
profile match exactly where your
elementary_test_results
table is being updated. This is a common source of confusion, especially when new datasets or environments are introduced or when switching between dev/prod targets [see example](https://slack.com/archives/C02CTC89LAX/p1659002103951729), [another example](https://slack.com/archives/C02CTC89LAX/p1739540598495729). 2. Target Consistency: Both your
dbt test
and
edr send-report
commands should use the same target/environment. If you run
dbt test
with
--target prod
, make sure the
elementary
profile used by the CLI is also configured for the
prod
environment. If there is a mismatch, the CLI may look in the wrong place for test results [see here](https://slack.com/archives/C02CTC89LAX/p1683764076675589). 3. Elementary Profile Name: The CLI specifically looks for a profile named
elementary
in your
profiles.yml
. This profile should be configured to point to the correct schema/database where the results are stored [see here](https://slack.com/archives/C02CTC89LAX/p1659002103951729). 4. Artifacts Location: If you have models and artifacts (like
elementary_test_results
) in different databases or schemas, Elementary may not be able to correlate them. The artifacts (Elementary tables) need to be in the same database as the models being tested for the hooks and reporting to work correctly [see here](https://slack.com/archives/C02CTC89LAX/p1733351444752069). What you should check: - Verify that the
elementary
profile in your
profiles.yml
is pointing to the same database and schema as your new dataset where the tests are being written. - Ensure you are not accidentally using a different profile or target between your dbt and Elementary CLI commands. - If you recently changed the schema or created a new dataset, update your
elementary
profile accordingly. If after these checks the issue persists, it may help to run
edr monitor debug
to validate the CLI's connection and configuration, as suggested in the sources. If you need more specific troubleshooting, sharing your
profiles.yml
(with sensitive info redacted) and the output of
edr monitor debug
can help pinpoint the issue. Let me know if you need help reviewing your profile configuration!
Sorry @Gaby Michel, I'm not sure what you mean. Please try again. 👋
g
@kapa.ai My profiles elementary: outputs: prod: type: bigquery method: oauth # or service-account project: data-analytics-414419 dataset: elementary # Elementary schema name threads: 8 timeout_seconds: 900 location: northamerica-northeast2 priority: interactive (airflow) π airflow/src/plat/dbt 287-data-quality-tests-cli-tool-and-mcp ✗ ❯ edr monitor debug --profiles-dir . ________ __ / ____/ /__ __ _ _ __ / /_____ ________ __ / __/ / / _ \/ _ `_ \/ _ \/ __ \/ _/ _ `/ ___/ / / / / /___/ / __/ / / / / / __/ / / / /_/ /_/ / / / /_/ / /_____/_/\___/_/ /_/ /_/\___/_/ /_/\__/\__,_/_/ \__, / /____/ Any feedback and suggestions are welcomed! join our community here - https://bit.ly/slack-elementary 2025-11-22 151905 — INFO — Running with edr=0.20.0 Connected to the Elementary db and schema successfully The bulk of my data is in one project, I store elementary data in another. Most of my pre existing tests show up fine, it's just the ones in the new dataset (inside my main project) 05_DQ are not showing up
k
A message from kapa.ai
Thank you for providing your
profiles.yml
and the debug output. Based on your setup and the knowledge sources, here’s what stands out: ### Key Points from Your Setup - Your
elementary
profile is configured to store Elementary data in the
elementary
dataset of the
data-analytics-414419
project. - You mention that most pre-existing tests show up fine, but tests for models in the new dataset (
05_DQ
in your main project) do not appear in the report, even though you see their results in the
elementary.elementary_test_results
table. ### What the Knowledge Sources Say 1. Elementary CLI Reads from a Single Schema/Project The Elementary CLI (
edr report
or
edr send-report
) will only read test results from the schema and project specified in the
elementary
profile. If your new models/tests are writing results to a different schema or project, the CLI will not pick them up unless the profile is updated to point there. However, you state that the test results for the new models do appear in the
elementary.elementary_test_results
table, which suggests the data is in the right place, but not showing up in the report. 2. Models and Test Results Must Be Correlated If models are in a different dataset/project than where Elementary is configured, the report may not be able to correlate test results with the models, especially if the model metadata (from
manifest.json
or dbt artifacts) does not match the location of the test results. This is a known limitation when using multiple datasets/projects, as Elementary expects both the model metadata and test results to be in the same logical space. This is discussed in several threads, where users with models in one project and Elementary data in another encounter issues with missing test results in the report, even though the data exists in the warehouse [see here](https://slack.com/archives/C02CTC89LAX/p1701347485583669), [and here](https://slack.com/archives/C02CTC89LAX/p1700170920246419). 3. Custom Datasets and Model Configuration If you have models in a custom dataset (like
05_DQ
), but your
elementary
profile and dbt project are not configured to recognize this, Elementary may not display those models/tests in the report. You may need to ensure that your
dbt_project.yml
and your test definitions are correctly referencing the new dataset, and that the Elementary dbt package is aware of it [see here](https://slack.com/archives/C02CTC89LAX/p1681162450753059). ### What to Check - Model Metadata Consistency: Ensure that the models in your new dataset (
05_DQ
) are being picked up by dbt and that their metadata (in
manifest.json
and in the warehouse) matches what Elementary expects. If the models are not present in the manifest or are not referenced correctly, their tests may not show up in the report. - Elementary Profile and dbt Config: Double-check that your
elementary
profile and your
dbt_project.yml
are configured so that both the models and the test results are associated with the same project and schema from the perspective of Elementary. If you have models in
05_DQ
, but the Elementary dbt package is not configured to look there, this could cause the issue. - Limitations with Multiple Projects/Datasets: Elementary currently does not fully support aggregating test results and model metadata across multiple projects or datasets into a single report. If your use case involves models in one project/dataset and Elementary data in another, this is a known limitation, and the recommended workaround is to keep both in the same project/dataset or generate separate reports for each [see here](https://slack.com/archives/C02CTC89LAX/p1701347485583669), [and here](https://slack.com/archives/C02CTC89LAX/p1681162450753059). ### Summary - If your new models/tests are in a different dataset or project than what the
elementary
profile points to, they will not show up in the report. - Even if the test results are in the correct table, if the model metadata is not aligned (i.e., the models are not recognized by Elementary in the same schema/project), the report will not display them. - This is a known limitation, and the best practice is to ensure both your models and Elementary data are in the same project/dataset, or to generate separate reports for each environment. If you need to support multiple datasets/projects, you may need to wait for improved multi-project support or adjust your architecture to fit within these constraints. If you have further details (such as your
dbt_project.yml
or how the new dataset is configured), I can help review those for alignment issues. --- References: - [Elementary profile and schema alignment](https://slack.com/archives/C02CTC89LAX/p1659002103951729) - [Known limitations with multiple projects/datasets](https://slack.com/archives/C02CTC89LAX/p1701347485583669) - [Custom dataset configuration](https://slack.com/archives/C02CTC89LAX/p1681162450753059)