Nodes both initialize and add computational resources to your Docker environment. They are an essential part of ensuring your clusters have enough power to meet the demand of your users and applications.

Overview

Specifications

🚧

Building a production environment?

Recommended: The D-series virtual machine sizes offer a combination of vCPU, memory recommended for most production workloads.

The B-series virtual machine sizes (Small and Medium) are ideal for workloads that do not need the full performance of the CPU continuously, like web servers, proof of concepts, small container databases and staging or development build environments.

Security

Networking

Nodes are automatically provisioned with communication privileges with other nodes and resources in the cluster. However, it is important to note that MedStack Control has been designed to disable connection attempts directly to nodes. This means that you cannot SSH into a node.

Encryption in transit

Data in transit between nodes is encompassed within the scope of Docker's overlay network.

Encryption at rest

All node disks are encrypted at rest to AES-256.

Intrusion detection system

All nodes are installed with the Threat Stack intrusion detection system (IDS).

Events detected by the IDS are managed and responded to by MedStack. MedStack adheres to our policies and controls available to you in the MedStack Control dashboard that explain how MedStack responds in the case of IDS events.

Actions

Create

You can create as many nodes as needed to power your clusters. Nodes can be created in just a few clicks.

1. In the Nodes section within the cluster, click the Create node or Create one now button.

2. Select the size of node you'd like to create and click the Create button. The size description expresses the number of cores in terms of vCPUs and the memory resources in terms of gigabytes.

3. You'll see the node status as Creating while it is being provisioned by the cloud provider. This can take up to 10 minutes.

4. Create additional nodes as your environment needs. Any node created after the Master (initial) node is designated as a Worker node.

View metrics

You can see resource consumption metrics for a node by clicking the View metrics → button from the actions dropdown button on a node. This will open a time-series graph to show CPU, Memory, and Storage utilization over time.

Labels

You can add labels to a node by clicking the Labels button from the actions dropdown button on a node.

You may use labels to organize the nodes within a cluster. Labels are key-value pairs that can be used to set up service placement constraints that control which nodes your containers will run on.

More information on using labels and placement constraints can be found on Docker's support documentation:

• Docker docs: Add or remove label metadata
• Docker docs: Using constraints and labels

❗️

Resizing the manager node

Be aware than when you resize the manager node, you will cause a brief disruption to your application's availability elapsing the time taken to scale the node on the cloud provider. This can take up to 5 minutes.

Reboot

You can reboot a node by clicking the Reboot button from the actions dropdown button on a node.

You may wish to reboot a node as a troubleshooting effort in the following events:

• The CPU or memory resources are exhausted.
• The status of a node is unavailable.

📘

Restart Docker

If your Docker state is not available, it's possible that you may need to restart Docker. You can do this by rebooting the manager node of the cluster.

Resize

You can resize a node by clicking the Resize button from the actions dropdown button on a node.

Remove stopped containers

Stopped containers can be removed from a node by selecting the Remove stopped containers button from the actions dropdown button on a node.

Removing stopped containers on a node permanently deletes stopped containers on that node. Other nodes with stopped containers will not be impacted by this action and will remain stopped on the other nodes.

By removing stopped containers, the logs and volume mounts configured in these containers will also be permanently deleted. The volumes attached by the mounts will not be deleted automatically by this action.

👍

Recently stopped containers with Exit errors

When you select the action to remove stopped containers, stopped containers within the past 24 hours that stopped due to an Exit error will not be removed so its logs can be viewed for troubleshooting purposes.

Perform maintenance

When new maintenance is available on a node, the Actions button is emphasized with a notification icon. Pending maintenances can be viewed by clicking See maintenance and can be started on demand.

Delete

You can delete a node by clicking the Delete button from the actions dropdown button on a node.

When you delete a node, the containers running on the node will be balanced across the remaining available nodes in the cluster while following the rules of any service placement constraints.

❗️

You'll delete data and unassign the static IP

When you delete a node, you also delete the disk mapping and the disk's data. Make sure you have moved important data from the node's disk that is to be deleted to another node in the cluster. You'll also lose the static IP address assigned to that node. There is no guarantee that creating another node will have the same static IP address.