David Mills
07/24/2025, 11:00 AMedr send-report from a GitHub action, to send a report in Slack. The full command I'm using is edr send-report --select last_invocation --days-back 7 --exclude-elementary-models true --project-dir . --profiles-dir ${{ env.DBT_PROFILES_DIR }} --profile-target default --env prod --slack-token ${{ env.SLACK_TOKEN }} --slack-channel-name ${{ env.SLACK_CHANNEL }} . The report is successfully being generated, uploaded to Slack and a summary message is sent. However, the number of tests run/passed/failed is different between the report UI and the slack summary message. See screenshots. What could be causing this?kapa.ai
07/24/2025, 11:01 AMDavid Mills
07/24/2025, 11:03 AMedr send-report from a GitHub action, to send a report in Slack. The full command I'm using is edr send-report --select last_invocation --days-back 7 --exclude-elementary-models true --project-dir . --profiles-dir ${{ env.DBT_PROFILES_DIR }} --profile-target default --env prod --slack-token ${{ env.SLACK_TOKEN }} --slack-channel-name ${{ env.SLACK_CHANNEL }} . The report is successfully being generated, uploaded to Slack and a summary message is sent. However, the number of tests run/passed/failed is different between the report UI and the slack summary message. What could be causing this?
The slack message says 337 passed, 89 warning, 0 failed.
The elementary report UI says:
• dbt tests: 330 passed, 83 warned, 0 failed
• freshness: 34 passed, 0 warned, 0 failedkapa.ai
07/24/2025, 11:03 AMkapa.ai
07/24/2025, 11:04 AM--select invocation_id:... or similar filters, and instead summarizes all recent test results, not just those for the selected invocation. This can cause the Slack summary to show numbers that do not match the filtered report UI. The Elementary team has acknowledged this and suggested opening a GitHub issue for tracking [Slack thread: summary ignores invocation filter](https://slack.com/archives/C02CTC89LAX/p1736401455291439). 3. Freshness Tests Handling Freshness tests are sometimes handled differently in the summary versus the report UI. For example, the report UI may break out freshness tests separately, while the Slack summary may aggregate them with other test types, or not include them at all, depending on the version and configuration [Slack thread: source freshness in report](https://slack.com/archives/C02CTC89LAX/p1689088826025229). What you can do: - Be aware that, as of now, the Slack summary may not always match the report UI due to the above issues. - If you need the summary to match the report UI exactly, you may need to wait for a fix or workaround from the Elementary team. They have indicated that improvements to summary generation and filtering are in progress [Slack thread: summary ignores invocation filter](https://slack.com/archives/C02CTC89LAX/p1736401455291439). - You can open or follow issues on the [Elementary GitHub](https://github.com/elementary-data/elementary/issues) for updates. Summary: The mismatch is most likely due to the Slack summary including skipped tests or not respecting invocation/tag filters, while the report UI shows only the filtered, actual test results. This is a known issue and is being addressed by the Elementary team. If you need more precise troubleshooting, sharing your edr.log and elementary_output.json files with the Elementary support team can help them diagnose your specific case [Slack thread: skipped tests in summary](https://slack.com/archives/C02CTC89LAX/p1699281535566889).