#C01AHCD885S Postgres syncing error: Saved offset is before replication slot's confirmed lsn. Please reset the connection, and then increase WAL retention and/or increase sync frequency to prevent this from happening in the future. See https://docs.airbyte.com/integrations/sources/postgres/postgres-troubleshooting#under-cdc-incremental-mode-there-are-still-full-refresh-syncs for more details. It does not look to me like the current lag value is above the wal_keep_size value set

This error ("Saved offset is before replication slot's confirmed lsn") can occur even when WAL retention settings appear sufficient. There are a few known causes and workarounds: Known Root Causes 1. PgJDBC Regression (Connector versions ≥ 3.6.19) A regression in PgJDBC 42.7.0+ causes

confirmed_flush_lsn

to automatically advance even when there are no new events for the publication tables — meaning the replication slot's LSN moves ahead of Airbyte's saved state, triggering this error. This is a known bug affecting connector versions 3.6.19 through at least 3.7.0. [CDC LSN issue] 2. Low Transaction Volume / WAL Disk Consumption On databases with low write activity, the WAL log sequence number may not advance naturally, causing similar LSN mismatches. [WAL disk consumption] ------------------------- Recommended Fixes Option A: Upgrade to connector version 3.7.2+ A fix was released in version

3.7.2

. Users have confirmed this resolves the issue. [CDC LSN issue] Option B: Downgrade to connector version 3.6.18 or 3.6.15 (temporary workaround) If upgrading isn't immediately possible, reverting to

3.6.18

or

3.6.15

has been confirmed to avoid the bug. [CDC LSN issue] Option C: Set up a heartbeat table This helps keep the LSN advancing on low-traffic databases: 1. Create the heartbeat table:

CREATE TABLE public.airbyte_heartbeat (
    id SERIAL PRIMARY KEY,
    timestamp TIMESTAMP NOT NULL DEFAULT current_timestamp,
    text TEXT
);

2. Add it to your publication:

ALTER PUBLICATION <publicationName> ADD TABLE public.airbyte_heartbeat;

3. In the Postgres source connector UI, under Advanced → Debezium heartbeat query, enter:

INSERT INTO public.airbyte_heartbeat (text) VALUES ('heartbeat');

[WAL disk consumption; CDC LSN issue] Option D: Change "Invalid CDC position behavior" to "Re-sync data" In the connector's advanced settings, switching from Fail sync to Re-sync data will allow the sync to recover via a full refresh, though this is not ideal for large tables and the error may recur. [CDC LSN issue] ------------------------- Summary: Since your WAL retention settings appear fine, the most likely culprit is the PgJDBC regression introduced in connector versions ≥ 3.6.19. The cleanest fix is to upgrade to connector version 3.7.2+, or use the heartbeat table workaround in the meantime. 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

wow major bug, upgrading the connector fixed it