> ## Documentation Index
> Fetch the complete documentation index at: https://docs.risingwave.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Streaming source offline

> Locate an offline source and restore ingestion safely.

This warning indicates that a streaming source has reported an offline status
for 15 minutes. Ingestion can be interrupted and downstream results can become
stale. An idle upstream with no new records is different from an offline source.

## Find the source and assess the impact

1. Open the affected cluster in the [Cloud dashboard](https://cloud.risingwave.com/dashboard).
2. Go to **Data Catalog** > **Sources** and open the source named in the alert.
   Use the source ID from the alert to distinguish it if needed.
3. Check its running status, errors, and throughput. Set the time range to
   include the start of the interruption. If you cannot access these details,
   ask your cluster administrator or Cloud support for help.
4. Check whether applications that depend on the source are receiving fresh
   results. Compare other sources to determine whether they share the failure.

## Restore ingestion

* For connection errors, check the upstream service's availability, network
  access, and connection limits.
* For authentication or permission errors, check the credentials and connector
  permissions, including any recent credential rotation.
* For a CDC source whose error mentions an unavailable replication position,
  check the connector-specific requirements for retained logs. See
  [MySQL CDC](https://docs.risingwave.com/ingestion/sources/mysql/mysql-cdc.md) or
  [PostgreSQL CDC](https://docs.risingwave.com/ingestion/sources/postgresql/pg-cdc.md), as applicable. Preserve
  required binlogs or WAL while investigating; consult support before resetting
  a replication position, recreating the source, or skipping records.

See [Source and sink troubleshooting](https://docs.risingwave.com/troubleshoot/troubleshoot-source-sink.md)
for additional diagnosis. If the upstream is healthy but the source stays
offline, the error is unclear, or production ingestion is affected, contact the
[RisingWave Cloud support team](mailto:cloud-support@risingwave-labs.com).
Include the cluster and region, source name and ID, upstream system type,
alert time, and representative error messages with credentials removed.

## Confirm recovery

Confirm that the source reports an online status, errors stop, and throughput
resumes when the upstream produces data. For CDC sources, use
[Monitor CDC progress](https://docs.risingwave.com/ingestion/monitor-cdc-progress.md) to check that ingestion
is catching up. Confirm downstream freshness as well. Zero throughput from an
idle upstream or a cleared alert alone is not proof of recovery.

## Applicability

This alert applies to clusters running RisingWave v1.7.1 or later.

## Related topics

* [Manage sources](https://docs.risingwave.com/cloud/manage-sources.md)
* [Source error](https://docs.risingwave.com/cloud/alert-playbooks/streaming/source-error.md)
* [Monitor messages and alerts](https://docs.risingwave.com/cloud/monitor-messages-and-alerts.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.