github.com/jmbataller/terraform@v0.6.8-0.20151125192640-b7a12e3a580c/website/source/docs/providers/aws/r/autoscaling_group.html.markdown (about) 1 --- 2 layout: "aws" 3 page_title: "AWS: aws_autoscaling_group" 4 sidebar_current: "docs-aws-resource-autoscaling-group" 5 description: |- 6 Provides an AutoScaling Group resource. 7 --- 8 9 # aws\_autoscaling\_group 10 11 Provides an AutoScaling Group resource. 12 13 ## Example Usage 14 15 ``` 16 resource "aws_autoscaling_group" "bar" { 17 availability_zones = ["us-east-1a"] 18 name = "foobar3-terraform-test" 19 max_size = 5 20 min_size = 2 21 health_check_grace_period = 300 22 health_check_type = "ELB" 23 desired_capacity = 4 24 force_delete = true 25 launch_configuration = "${aws_launch_configuration.foobar.name}" 26 27 tag { 28 key = "foo" 29 value = "bar" 30 propagate_at_launch = true 31 } 32 tag { 33 key = "lorem" 34 value = "ipsum" 35 propagate_at_launch = false 36 } 37 } 38 ``` 39 40 ## Argument Reference 41 42 The following arguments are supported: 43 44 * `name` - (Optional) The name of the auto scale group. By default generated by terraform. 45 * `max_size` - (Required) The maximum size of the auto scale group. 46 * `min_size` - (Required) The minimum size of the auto scale group. 47 (See also [Waiting for Capacity](#waiting-for-capacity) below.) 48 * `availability_zones` - (Optional) A list of AZs to launch resources in. 49 Required only if you do not specify any `vpc_zone_identifier` 50 * `launch_configuration` - (Required) The name of the launch configuration to use. 51 * `health_check_grace_period` - (Optional) Time after instance comes into service before checking health. 52 * `health_check_type` - (Optional) "EC2" or "ELB". Controls how health checking is done. 53 * `desired_capacity` - (Optional) The number of Amazon EC2 instances that 54 should be running in the group. (See also [Waiting for 55 Capacity](#waiting-for-capacity) below.) 56 * `min_elb_capacity` - (Optional) Setting this will cause Terraform to wait 57 for this number of healthy instances all attached load balancers. 58 (See also [Waiting for Capacity](#waiting-for-capacity) below.) 59 * `force_delete` - (Optional) Allows deleting the autoscaling group without waiting 60 for all instances in the pool to terminate. You can force an autoscaling group to delete 61 even if it's in the process of scaling a resource. Normally, Terraform 62 drains all the instances before deleting the group. This bypasses that 63 behavior and potentially leaves resources dangling. 64 * `load_balancers` (Optional) A list of load balancer names to add to the autoscaling 65 group names. 66 * `vpc_zone_identifier` (Optional) A list of subnet IDs to launch resources in. 67 * `termination_policies` (Optional) A list of policies to decide how the instances in the auto scale group should be terminated. 68 * `tag` (Optional) A list of tag blocks. Tags documented below. 69 * `wait_for_capacity_timeout` (Default: "10m") A maximum 70 [duration](https://golang.org/pkg/time/#ParseDuration) that Terraform should 71 wait for ASG instances to be healthy before timing out. (See also [Waiting 72 for Capacity](#waiting-for-capacity) below.) Setting this to "0" causes 73 Terraform to skip all Capacity Waiting behavior. 74 75 Tags support the following: 76 77 * `key` - (Required) Key 78 * `value` - (Required) Value 79 * `propagate_at_launch` - (Required) Enables propagation of the tag to 80 Amazon EC2 instances launched via this ASG 81 82 ## Attributes Reference 83 84 The following attributes are exported: 85 86 * `id` - The autoscaling group name. 87 * `availability_zones` - The availability zones of the autoscale group. 88 * `min_size` - The minimum size of the autoscale group 89 * `max_size` - The maximum size of the autoscale group 90 * `default_cooldown` - Time between a scaling activity and the succeeding scaling activity. 91 * `name` - The name of the autoscale group 92 * `health_check_grace_period` - Time after instance comes into service before checking health. 93 * `health_check_type` - "EC2" or "ELB". Controls how health checking is done. 94 * `desired_capacity` -The number of Amazon EC2 instances that should be running in the group. 95 * `launch_configuration` - The launch configuration of the autoscale group 96 * `vpc_zone_identifier` - The VPC zone identifier 97 * `load_balancers` (Optional) The load balancer names associated with the 98 autoscaling group. 99 100 ~> **NOTE:** When using `ELB` as the health_check_type, `health_check_grace_period` is required. 101 102 <a id="waiting-for-capacity"></a> 103 ## Waiting for Capacity 104 105 A newly-created ASG is initially empty and begins to scale to `min_size` (or 106 `desired_capacity`, if specified) by launching instances using the provided 107 Launch Configuration. These instances take time to launch and boot. 108 109 Terraform provides two mechanisms to help consistently manage ASG scale up 110 time across dependent resources. 111 112 #### Waiting for ASG Capacity 113 114 The first is default behavior. Terraform waits after ASG creation for 115 `min_size` (or `desired_capacity`, if specified) healthy instances to show up 116 in the ASG before continuing. 117 118 Terraform considers an instance "healthy" when the ASG reports `HealthStatus: 119 "Healthy"` and `LifecycleState: "InService"`. See the [AWS AutoScaling 120 Docs](https://docs.aws.amazon.com/AutoScaling/latest/DeveloperGuide/AutoScalingGroupLifecycle.html) 121 for more information on an ASG's lifecycle. 122 123 Terraform will wait for healthy instances for up to 124 `wait_for_capacity_timeout`. If ASG creation is taking more than a few minutes, 125 it's worth investigating for scaling activity errors, which can be caused by 126 problems with the selected Launch Configuration. 127 128 Setting `wait_for_capacity_timeout` to `"0"` disables ASG Capacity waiting. 129 130 #### Waiting for ELB Capacity 131 132 The second mechanism is optional, and affects ASGs with attached Load 133 Balancers. If `min_elb_capacity` is set, Terraform will wait for that number of 134 Instances to be `"InService"` in all attached `load_balancers`. This can be 135 used to ensure that service is being provided before Terraform moves on. 136 137 As with ASG Capacity, Terraform will wait for up to `wait_for_capacity_timeout` 138 (for `"InService"` instances. If ASG creation takes more than a few minutes, 139 this could indicate one of a number of configuration problems. See the [AWS 140 Docs on Load Balancer 141 Troubleshooting](https://docs.aws.amazon.com/ElasticLoadBalancing/latest/DeveloperGuide/elb-troubleshooting.html) 142 for more information.