We have a cluster version 1.25 with several cce_node_pools. In the console we see a notification about the opportunity to update the cluster to version 1.27, but the update check fails due to the fact that cce_node_pools are deployed with docker support, and version 1.27 only supports containerd.
At the same time, even when deploying a pool for a cluster version 1.25 from scratch, nodes are created with the docker containerization mechanism, which does not make it possible to smoothly transfer all exploited resources to the new version of the cluster.
We found an opportunity to manually create a cce_node_pool with containerd support via the console, send the nodes of the old pool to drain and thus bring the cluster to a state from which it can be updated.
But taking into account the need to subsequently bring the state up to date, this takes a decent amount of time.
How can we specify the containerization engine directly when deploying a pool using terraform?
We have a cluster version 1.25 with several cce_node_pools. In the console we see a notification about the opportunity to update the cluster to version 1.27, but the update check fails due to the fact that cce_node_pools are deployed with docker support, and version 1.27 only supports containerd. At the same time, even when deploying a pool for a cluster version 1.25 from scratch, nodes are created with the docker containerization mechanism, which does not make it possible to smoothly transfer all exploited resources to the new version of the cluster.
We found an opportunity to manually create a cce_node_pool with containerd support via the console, send the nodes of the old pool to drain and thus bring the cluster to a state from which it can be updated. But taking into account the need to subsequently bring the state up to date, this takes a decent amount of time.
How can we specify the containerization engine directly when deploying a pool using terraform?