Affected Time

February 25, 2026, 14:00 – 20:00 (UTC+8)

Regions: Singapore, Zhangjiakou, Shenzhen, Hong Kong, US (Silicon Valley), US (Virginia), Germany (Frankfurt), Indonesia (Jakarta)

February 26, 2026, 14:00 – 20:00 (UTC+8)

Region: Hangzhou

February 27, 2026, 14:00 – 20:00 (UTC+8)

Region: Beijing

February 28, 2026, 14:00 – 20:00 (UTC+8)

Region: Shanghai

Change Details

The underlying environment of Realtime Compute for Flink will be upgraded.

Impact

During the upgrade, some jobs may experience failover. After failover, the system will attempt to recover jobs based on their configured restart strategies.

We strongly recommend setting your job’s restart strategy to "restart". Jobs with restart strategy set to "none" will not be automatically recovered after failover.

The console and API services will remain available during the upgrade and will not be affected. You may also proactively check and restart affected jobs via the console in advance to minimize potential impact.

We apologize for any inconvenience caused. If you have any questions or encounter issues, please feel free to contact us through support tickets.

Let me know if you'd like a shorter version or need it formatted for email or official announcement.