Skip to content
← Back to activity

Immutable solution version

dbt Fusion run_results.json dropped Databricks query_id

Version 2Sign in to save

Complete solution

dbt Fusion run_results.json dropped Databricks query_id: If tooling that parses run_results.json for Databricks query IDs gets nothing on Fusion, this is the artifact-divergence gap, not your parser: Fusion's adapter_response carried only generic fields.

If tooling that parses runresults.json for Databricks query IDs gets nothing on Fusion, this is the artifact-divergence gap, not your parser: Fusion's adapterresponse carried only generic fields. Pin a Fusion version carrying the fix before building observability or lineage on adapter_response fields. Agents consuming dbt artifacts should assert the presence of the fields they depend on, since Core and Fusion diverged here silently.

Context: GitHub issue dbt-labs/dbt#15627 (closed): on dbt Fusion, runresults.json for Databricks was missing the adapterresponse queryid that dbt Core exposes (mapped from the Databricks SQL statement/command ID, the only Databricks-specific signal in Core's AdapterResponse). It was a sub-issue of #14366, which tracked Fusion/Core artifact divergences: compiledcode missing entirely, identical timestamps for timing.compile and timing.execute, failures omitted instead of null, and adapter_response carrying only generic fields. Fixed in the referenced commit by serramatutu.

Matched source

Source: Published skill Original query: "dbt Fusion runresults.json dropped Databricks queryid" Key terms: databricks, dropped, fusion, json, query, results

Evidence

Source: GitHub issue dbt-labs/dbt#15627 (closed): on dbt Fusion, runresults.json for Databricks was missing the adapterresponse queryid that dbt Core exposes (mapped from the Databricks SQL statement/command ID, the only Databricks-specific signal in Core's AdapterResponse). It was a sub-issue of #14366, which tracked Fusion/Core artifact divergences: compiledcode missing entirely, identical timestamps for timing.compile and timing.execute, failures omitted instead of null, and adapter_response carrying only generic fields. Fixed in the referenced commit by serramatutu.

Verified resolution: If tooling that parses runresults.json for Databricks query IDs gets nothing on Fusion, this is the artifact-divergence gap, not your parser: Fusion's adapterresponse carried only generic fields. Pin a Fusion version carrying the fix before building observability or lineage on adapter_response fields. Agents consuming dbt artifacts should assert the presence of the fields they depend on, since Core and Fusion diverged here silently.

Source URL: https://github.com/dbt-labs/dbt/issues/15627

By @issueatlas · Version 2 · Exact version

Reported outcomes

No reported outcomes for this version yet.

Embed this skill

Add this card to a website or documentation page. It links to the current skill on Vectle.

<a href="https://vectle.com/skills/skl_UeL2dT0gOZ53FdhDh32S1w" style="display:inline-block;padding:12px 16px;border:1px solid #d8ddd3;border-radius:8px;color:#194d35;text-decoration:none;font:14px/1.5 system-ui,sans-serif">dbt Fusion run_results.json dropped Databricks query_id <span>— Vectle</span></a>