Can a non-highly available cluster be upgraded to a highly available cluster?
Due to the significant differences in deployment architecture between non-highly available clusters and highly available clusters, it is not possible to convert between these two deployment forms. If adjustments are necessary, please repurchase the appropriate deployment form. Note that non-highly available clusters are only for testing purposes and should not be used in a production environment.
What are the available components and their respective versions?
EMR regularly releases new product versions. The components and versions currently available for purchase can be viewed in the Product Versions section.
After the cluster is created, is it possible to upgrade the component version?
This is not supported.
After the cluster is created, is it possible to add components?
Components can be added, with the selection range being the undeployed components currently supported by the cluster product version.
After the cluster is created, is it possible to modify the project?
Modifying the project to which the cluster belongs is not supported. It is recommended to use the tag function instead of the project capability. The tag function supports tag authentication, tag accounting, and other capabilities.
Can different machine specifications be selected for nodes of the same type?
Nodes of the same type deploy consistent services. For ease of operation and management, it is recommended that nodes of the same type use the same machine specifications. If the machine specifications are sold out or do not meet current requirements, the machine specifications can be modified for expansion.
Is it possible to reduce the configuration of existing nodes?
In the EMR on CVM scenario, a variety of services are deployed on each node type, sharing the same node's computing, memory, and disk resources. If the node's physical configuration is reduced, it is easy for the total resource usage in the actual service configuration file to exceed the total physical resources of the node, which will render the cluster unusable. Therefore, the function to reduce the configuration of existing nodes is not provided. If the current configuration of Core nodes, TASK nodes, and Router nodes is too high and there is a lot of surplus, you can expand nodes with lower configurations first and then reduce nodes with higher configurations, achieving a reduction through node replacement.
Is it possible to reduce the capacity of Core nodes?
Core nodes are primarily used for data storage and by default, do not support reduction in capacity. If a reduction is indeed necessary, please contact us through a work order.