High Risk Operations

Edit

KubeSphere supports high availability of master and working nodes to ensure cluster sustainability and stability owing to the underlying kubernetes scheduling mechanism. It is not recommended to shut down or restart nodes for no reason, because such operations are high-risk that may cause the services unavailable. The following operations are the ones we can think of people should pay much attention to.

High-risk Operations

SequenceHigh-risk Operation
1Restart the cluster node or reinstall the operating system.
2It is recommended that you do not install additional software on the cluster node, which could make the cluster node unavailable.

Prohibitive Operations

SequenceProhibitive Operations
1Delete /var/lib/etcd/. Delete /var/lib/docker. Delete /etc/kubernetes/. Delete /etc/kubesphere/
2Disk formatting, partitioning.