# External data sync failures - Docs

The external data sync failures check fires when one of your [external data sources](/docs/cdp/sources.md) fails to sync, which means the connected data in your warehouse is stale or incomplete.

![External data sync failures check in the Health checks section in the PostHog app](https://res.cloudinary.com/dmukukwp6/image/upload/w_1000,c_limit,q_auto,f_auto/pasted_image_2026_06_24_T20_15_15_848_Z_05efe94e94.png)![External data sync failures check in the Health checks section in the PostHog app](https://res.cloudinary.com/dmukukwp6/image/upload/w_1000,c_limit,q_auto,f_auto/pasted_image_2026_06_24_T20_15_29_332_Z_bd8ba04ac5.png)

## What this check looks for

Once a day, the check looks at your external data schemas and raises a **warning** for any that have failed, hit a billing limit, or recorded an error and stopped syncing. The issue includes the source name, source type, and the error from the failed run.

## Why it matters

The connected data is stale or incomplete until the sync succeeds again. The most common causes are:

-   Expired or rotated credentials.
-   A permissions change on the source.
-   An upstream schema change – a renamed or removed column.

## Fix it manually

1.  Open the **Pipeline status** page (Data pipeline / Data Warehouse → Sources).
2.  Find the failing source, open its latest run, and read the error.
3.  Fix the root cause – reconnect or update the credentials, re-grant access, or update the schema mapping.
4.  Trigger a new sync and confirm it completes.

## This check is not auto-fixed

> Unlike the other checks, external data sync failures are **surfaced as alerts but do not open an automatic fix**. Fixing them usually requires re-entering credentials, which can't be done over the API – so this is the one check [PostHog Code](/docs/posthog-code.md) intentionally leaves to you.

That said, if you investigate it through PostHog Code, the agent can still diagnose the failure for you – listing and inspecting the source, reading the actual sync error, and applying config-only fixes (like resyncing or updating a schema mapping) where credentials aren't the problem. It will surface the exact error and ask you to reconnect the source when credentials are involved.

## Related

-   [Health checks overview](/docs/health-checks.md)
-   [Data Warehouse sources](/docs/cdp/sources.md)
-   [Materialized view failures](/docs/health-checks/materialized-view-failures.md)

### Community questions

Ask a question

### Was this page useful?

HelpfulCould be better