I have SQL Server as a source and Databricks as destination, the connection is configured correctly. I've configured the streams (tables) to do full refresh / overwrite. When i run the connection i get 0 records, even though if i use the same SQL user and query the same table i do get rows in SSMS, what can be the root cause? More details: The connection duration is about 20 minutes, stuck after the orchestration image is created. Airbyte internal table is created but no data inside

Based on the provided knowledge, there is a very similar case where: • Source: SQL Server • Destination: SQL Server • Behavior: ◦ Connection and discovery succeed ◦ Tables/schemas are created in the destination ◦ Sync runs for a long time (around 20–40 minutes) ◦ 0 records are written ◦ No obvious error in the UI/logs • Root cause: insufficient CPU resources in Kubernetes, which prevented the replication pod from starting properly. The logs contained: >

ResourceConstraintException: Unable to start the REPLICATION pod. This may be due to insufficient system resources. Please check available resources and try again.

[MSSQL→MSSQL issue] Once CPU resources were increased, the user confirmed: “Problem solved, It's about insufficient CPU in kubernetes.” [MSSQL→MSSQL issue] In your case (SQL Server → Databricks), the knowledge sources do not explicitly mention Databricks, but given: • The long run time (~20 minutes) • The job “stuck after the orchestration image is created” • Internal/Airbyte tables created but no data the closest documented root cause is the same: the replication pod is not actually starting due to resource constraints (CPU/memory) on the Airbyte/Kubernetes cluster, so no records are ever read/emitted, yet the setup phase (schema/table creation) can still occur. [MSSQL→MSSQL issue] What to check (based on the issue): 1. Inspect the job logs for messages like:

ResourceConstraintException: Unable to start the REPLICATION pod...

[MSSQL→MSSQL issue] 2. Verify that the Kubernetes cluster (or whatever orchestrator you use) has enough CPU/memory to schedule: ◦ The orchestrator container ◦ The source connector container ◦ The destination connector container [MSSQL→MSSQL issue] 3. If you see such errors, increase available resources or lower the resource requests/limits for the workload so the replication pod can start. If, after confirming resources are sufficient, you still see 0 records and no SELECT/EMITTED RECORDS in logs, the knowledge sources do not provide further Databricks-specific causes, so I can’t reliably point to another root cause beyond the documented resource-constraint issue. 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

what are the recommended specs for kubernetes on airbyte?