Rafael Felipe
04/22/2026, 12:31 PMkapa.ai
04/22/2026, 12:31 PMYYYY-MM-DD or YYYY-MM-DDTHH:MM:SSZ) that covers the range of data you need. [setup guide] 2. Object Not Queryable Airbyte can only fetch Salesforce objects that have the queryable property set to true. If the object is not queryable, Airbyte won't be able to retrieve data from it, even if the API user has access. [supported streams] 3. User Permissions Even if the same user can access data via the API directly, the Airbyte connector may be hitting a specific object or field permission issue. Ensure the authenticated user has:
⢠Read access to the object
⢠Field-level access to the relevant fields
⢠API access enabled [dedicated user] 4. Object Not Supported by BULK API Some Salesforce objects are not supported by the BULK API and fall back to the REST API. If the object has issues with the BULK API, it may silently fail. You can check if the object is in the list of known unsupported objects or try enabling Force Use Bulk API to see if behavior changes. [rate limiting] 5. QUERY_TIMEOUT Error Being Silently Swallowed There is a known issue where a QUERY_TIMEOUT error from Salesforce (which occurs when a query runs too long on large objects) may not be properly raised as an exception, causing the sync to report 0 records without a clear error. [QUERY_TIMEOUT forum] 6. Missing Records Due to API Eventual Consistency For incremental syncs, Salesforce does not guarantee immediate availability of recently created/updated records. If records are missing, try increasing the Lookback Window in the connector settings (e.g., from PT10M to PT30M or PT1H). [missing records] Recommended next steps:
1. Check and set an appropriate Start Date.
2. Verify the object is queryable and the user has full read permissions.
3. Review the sync logs in the Airbyte UI for any error messages (e.g., QUERY_TIMEOUT, FORBIDDEN, BAD_REQUEST). š 0 button š 0 button ā¹ļø Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.Rafael Felipe
04/22/2026, 12:37 PMkapa.ai
04/22/2026, 12:37 PMstream_slice_step controls how the time range is split into intervals for incremental syncs. With P60D (60-day intervals), the connector will request data in 60-day chunks between the start date and now.
From the source code, the default slice step is P30D (30 days). You've set it to P60D, which means fewer but larger requests. This is generally fine, but if the table has a large volume of records within a 60-day window, it could potentially cause query timeouts or other issues on the Salesforce side. What to Check 1. Set an explicit Start Date: Even if you want data from the last 2 years, setting an explicit start date ensures the connector behaves predictably. Use format YYYY-MM-DD or YYYY-MM-DDTHH:MM:SSZ. [setup guide]
2. Verify the cursor field: For incremental syncs, the connector queries using a cursor field (e.g., SystemModStamp). If the records in the table don't have values in that cursor field within the last 2 years, they won't be returned.
3. Check the Lookback Window: If records are being missed due to Salesforce API eventual consistency, increase the Lookback Window (default is PT10M). [missing records]
4. Try reducing the slice step: If the 60-day window is too large and causing timeouts, try reducing it back to P30D or even smaller (e.g., P7D) to see if records start appearing.
In summary, the combination of no start date (defaulting to 2 years back) and a P60D slice step should still pull data ā so if you're seeing 0 records, the issue is likely related to permissions, the cursor field values, or query timeouts rather than the slice step itself. š 0 button š 0 button ā¹ļø Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.