You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/create-cluster.md
+71-55Lines changed: 71 additions & 55 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,14 +3,13 @@
3
3
`eksdemo` can manage applications in any EKS cluster and the cluster doesn’t have to be created by `eksdemo`. You can use `eksctl` to create the cluster and then manage application using `eksdemo`. However, there are a number of benefits to using `eksdemo` to create your cluster:
4
4
* Cluster logging is enabled by default
5
5
* OIDC is enabled by default so IAM Roles for Service Accounts (IRSA) works out of the box
6
-
* The Managed Node Group ASG max is set to 10 so cluster autoscaling can work out of the box
6
+
* The Managed Node Group ASG max is set to 10 so Cluster Autoscaler can work out of the box
7
7
* Private networking for nodes is set by default
8
-
* VPC CNI is configured as a Managed Add-on and configured with IRSA by default
8
+
* VPC CNI is configured as a Managed Add-on and configured with IRSA, with network policy enabled
9
9
* t3.large instances used by default instead of m5.large for cost savings, but can be easily changed with the `--instance` flag or the shorthand `-i`
10
-
* To use containerd as the CRI on Amazon EKS optimized Amazon Linux AMIs is as easy as using the `--containerd` boolean flag
11
10
* To create a Fargate profile that selects workloads in the “fargate” namespace, use the `--fargate` boolean flag
12
-
* Choose a supported EKS version with the `--version` flag or the shorthand `-v` like `-v 1.21`
13
-
* Using a different OS like Bottlerocket or Ubuntu is as easy as `--os bottlerocket` or `--os ubuntu`
11
+
* Choose a supported EKS version with the `--version` flag or the shorthand `-v` like `-v 1.29`
12
+
* Using a different OS like Bottlerocket or AL2023 is as easy as `--os bottlerocket` or `--os amazonlinux2023`
14
13
* To use IPv6 networking, set the `--ipv6` boolean flag
15
14
* If you need to further customize the config, add the `--dry-run` flag and it will output the eksctl YAML config file and you can copy/paste it into a file, make your edits and run `eksctl create cluster -f cluster.yaml`
16
15
@@ -27,19 +26,26 @@ Aliases:
27
26
cluster, clusters
28
27
29
28
Flags:
30
-
--containerd use containerd runtime
31
-
--dry-run don't create, just print out all creation steps
32
-
--fargate create a Fargate profile
33
-
-h, --help help for cluster
34
-
-i, --instance string instance type (default "t3.large")
35
-
--ipv6 use IPv6 networking
36
-
--max int max nodes (default 10)
37
-
--min int min nodes
38
-
--no-roles don't create IAM roles
39
-
-N, --nodes int desired number of nodes (default 2)
40
-
--os string Operating System (default "AmazonLinux2")
41
-
--private private cluster (includes ECR, S3, and other VPC endpoints)
42
-
-v, --version string Kubernetes version (default "1.24")
29
+
--disable-network-policy don't enable network policy for Amazon VPC CNI
30
+
--dry-run don't create, just print out all creation steps
31
+
--encrypt-secrets string alias of KMS key to encrypt secrets
32
+
--fargate create a Fargate profile
33
+
-h, --help help for cluster
34
+
-H, --hostname-type string type of hostname to use for EC2 instances (default "resource-name")
35
+
-i, --instance string instance type (default "t3.large")
36
+
--ipv6 use IPv6 networking
37
+
--kubeconfig string path to write kubeconfig (default "/Users/jsmith/.kube/config")
38
+
--max int max nodes (default 10)
39
+
--min int min nodes
40
+
--no-roles don't create IAM roles
41
+
--no-taints don't taint nodes with GPUs or Neuron cores
42
+
-N, --nodes int desired number of nodes (default 2)
43
+
--os string Operating System (default "AmazonLinux2")
44
+
--prefix-assignment configure VPC CNI for prefix assignment
45
+
--private private cluster (includes ECR, S3, and other VPC endpoints)
46
+
-v, --version string Kubernetes version (default "1.30")
47
+
--vpc-cidr string CIDR to use for EKS Cluster VPC (default "192.168.0.0/16")
48
+
--zones strings list of AZs to use. ie. us-east-1a,us-east-1b,us-east-1c
43
49
44
50
Global Flags:
45
51
--profile string use the specific profile from your credential file
To view detail on the nodes, use the **`eksdemo get node`** command. Here we’ll use the `-c` flag which is the shorthand version of the `--cluster` flag.
0 commit comments