Scheduled Maintenance of Confluent Cloud
On August 31st, 2024, between 2:00 AM and 3:00 AM UTC (7:00 PM and 8:00 PM PST on August 30th, 2024) Confluent Cloud will undergo scheduled maintenance for an infrastructure upgrade. Once started, the maintenance will last approximately 1 hour. During this time, you can access your existing clusters, their metrics and the flow of data to and from your producers and consumers will be unaffected. However, you will not be able to use Confluent Cloud’s configuration and management services, such as creating new clusters or making configuration changes.
We appreciate your patience and understanding.
What to expect during the scheduled maintenance period
This maintenance impacts new deployments and configuration changes to your Confluent clusters and other services. During the maintenance period you will be unable to make changes to your Confluent Cloud environments. The scheduled maintenance will have no impact on any of your running Confluent Cloud components; clusters, Connect, ksqlDB, Kafka Streams, Flink and Schema Registry.
You can reduce the impact by doing the following before the maintenance window:
- Plan your management operations and configuration changes to occur before or after this maintenance window.
- If you have automation that relies on the availability of Confluent APIs, make sure you take steps to mitigate their unavailability during the maintenance window. All services and operations will automatically be available once the maintenance is complete.
- Have API keys ready ahead of time to get metrics from your clusters. You must obtain API keys before the maintenance window starts as you won’t be able to create API keys during maintenance. An API key is required to use the Metrics API.
During the maintenance window:
- Follow the Confluent status page for updates.
If you experience any issues outside of what has been described in this document, contact Confluent Support and open a ticket.
Details by scenario
The following table provides more details about what features are impacted and what is available for use during this scheduled maintenance window.
| Scenario | Impact | Mitigation |
| Create or delete a cluster | All of the CRUD (create, read, update, delete) operations for Basic, Standard, Enterprise and Dedicated clusters will not be available using any of the following access methods: | Plan your management and configuration operations around this maintenance window. Suspend automation that relies on API availability during the maintenance window. All services and operations will automatically be available once the maintenance is complete. |
| Access a cluster | Your clusters can be accessed through the connection end points and you will have complete connectivity to the cluster during the maintenance window | No mitigation required |
| Create or delete topics in a cluster | You can access CRUD operations for a topic with the Cluster administration APIs These APIs will be available and are not affected by the scheduled maintenance. | No mitigation required. |
| Client producers and consumers access to data | Your producers and consumers can stream data through your Kafka clusters and are not affected by the scheduled maintenance. | No mitigation required. |
| Cluster monitoring | The Metrics API will be available during the maintenance window to help you monitor the health and availability of your Kafka clusters. Note that Metrics will only be visible via APIs, not via the Cloud Console or CLI. | No mitigation required. |
| Add new roles | New roles and new role bindings cannot be added during the maintenance window. | Wait until after the maintenance completes to add new roles and role bindings. |
| Access to Confluent APIs |
|
Make sure any automation that relies on these APIs is paused for the maintenance window, or can gracefully fail. |
| Confluent CLI & TF | Confluent CLI & TF will not be available during the maintenance window. | Avoid using the CLI & TF during the maintenance window. |
| Deployment events during the maintenance window | You cannot deploy during the maintenance window. | Provisioning requests that are received during the maintenance window will be paused and will continue after the maintenance completes. Check the Confluent status page for updates. Any provisioning requests should be completed within a 24 hour period, regardless of the scheduled maintenance. Contact Confluent Support if provisioning takes longer than 24 hours. |
| ksqlDB, Connect, SR clusters and Flink Compute Pool provisioning | All provisioning services will be unavailable, meaning you cannot provision, delete or update your clusters. However, if you have already initiated CRUD operations, they will resume after the maintenance is complete. | Plan your management and configuration operations around this maintenance window. If you have automation that relies on the availability of ksqlDB, Connect, SR cluster or Compute Pool provisioning, make sure you take steps to mitigate their unavailability during the maintenance window. All services and operations will automatically be available once the maintenance is complete. |
| Monitor and manage connector status | Connector status will not be updated during the maintenance window | Plan your management and configuration operations around this maintenance window. Contact Confluent Support and open a ticket for remediation of connectors, if needed. |
| Purchase of Confluent Cloud on the AWS, Azure and GCP marketplaces | Purchase requests through cloud marketplaces will be queued and processed after the maintenance window closes. | You should refrain from purchasing PAYG and Commit subscriptions from all 3 cloud providers until after the scheduled maintenance is complete. |
| Access to Confluent Cloud through my AWS, Azure or GCP marketplace accounts | Sign in to Confluent through your marketplace accounts along with SSO will be unavailable during the scheduled maintenance window and will resume normally afterwards. | Avoid accessing your account through your marketplace accounts during the maintenance window. |