Maintenance Notice (September 7, 2026)

Scheduled maintenance is planned as follows. We apologize for any inconvenience and appreciate your understanding and cooperation.

Scheduled Maintenance Date:

Monday, September 7, 2026

Time:

9:00 – 18:00

Affected Service:

Kubernetes Service (including the Workspace feature)

Affected Sites:

All sites

Reason:

Impact:

As each compute node is upgraded, Pods running on the node being upgraded will be deleted.
Deleted Pods will be recreated on another compute node or the same compute node according to the rescheduling policy.
- Pods that are not managed by higher-level resources such as Deployments or Jobs (Bare Pods) will not be recreated automatically.
Runtime environments in the Workspace feature will also be temporarily stopped and recreated.
The impact on data retention is as follows.
- Data stored on persistent storage (such as PVCs) will be retained.
- Data in Pod or Workspace runtime environments that is not stored on persistent storage will be lost.
- Data in memory and the state of computations in progress will not be retained.
During maintenance, the Kubernetes API may become temporarily unstable, and operations using kubectl or the Workspace feature may fail.

Shared nodes will be upgraded sequentially, with up to 25% of the nodes (rounded down) temporarily unavailable at the same time.
All dedicated nodes allocated to users will also be upgraded sequentially.
- If fewer than 4 nodes are under contract, one node at a time will be temporarily unavailable.
- If 4 or more nodes are under contract, up to 25% of the nodes (rounded down) will be temporarily unavailable at the same time.
- If only one node is under contract, no dedicated node will be available temporarily while that node is being upgraded.

Please ensure that all critical data is saved to persistent storage before the maintenance begins.

Please refer to the following user guides as needed for information on how to save data, maintenance policies, and other details: