salt.states.boto_asg¶

Manage Autoscale Groups¶

New in version 2014.7.0.

Create and destroy autoscale groups. Be aware that this interacts with Amazon's services, and so may incur charges.

This module uses boto, which can be installed via package, or pip.

This module accepts explicit autoscale credentials but can also utilize IAM roles assigned to the instance trough Instance Profiles. Dynamic credentials are then automatically obtained from AWS API and no further configuration is necessary. More Information available at:

http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/iam-roles-for-amazon-ec2.html

If IAM roles are not used you need to specify them either in a pillar or in the minion's config file:

asg.keyid: GKTADJGHEIQSXMKKRBJ08H
asg.key: askdjghsdfjkghWupUjasdflkdfklgjsdfjajkghs

It's also possible to specify key, keyid and region via a profile, either as a passed in dict, or as a string to pull from pillars or minion config:

myprofile:
    keyid: GKTADJGHEIQSXMKKRBJ08H
    key: askdjghsdfjkghWupUjasdflkdfklgjsdfjajkghs
    region: us-east-1

Ensure myasg exists:
  boto_asg.present:
    - name: myasg
    - launch_config_name: mylc
    - availability_zones:
      - us-east-1a
      - us-east-1b
    - min_size: 1
    - max_size: 1
    - desired_capacity: 1
    - load_balancers:
      - myelb
    - suspended_processes:
        - AddToLoadBalancer
        - AlarmNotification
    - scaling_policies
        ----------
        - adjustment_type: ChangeInCapacity
        - as_name: api-production-iad
        - cooldown: 1800
        - min_adjustment_step: None
        - name: ScaleDown
        - scaling_adjustment: -1
    - region: us-east-1
    - keyid: GKTADJGHEIQSXMKKRBJ08H
    - key: askdjghsdfjkghWupUjasdflkdfklgjsdfjajkghs

# Using a profile from pillars.
Ensure myasg exists:
  boto_asg.present:
    - name: myasg
    - launch_config_name: mylc
    - availability_zones:
      - us-east-1a
      - us-east-1b
    - min_size: 1
    - max_size: 1
    - desired_capacity: 1
    - load_balancers:
      - myelb
    - profile: myprofile

# Passing in a profile.
Ensure myasg exists:
  boto_asg.present:
    - name: myasg
    - launch_config_name: mylc
    - availability_zones:
      - us-east-1a
      - us-east-1b
    - min_size: 1
    - max_size: 1
    - desired_capacity: 1
    - load_balancers:
      - myelb
    - profile:
        keyid: GKTADJGHEIQSXMKKRBJ08H
        key: askdjghsdfjkghWupUjasdflkdfklgjsdfjajkghs
        region: us-east-1

# Deleting an autoscale group with running instances.
Ensure myasg is deleted:
  boto_asg.absent:
    - name: myasg
    # If instances exist, we must force the deletion of the asg.
    - force: True

It's possible to specify cloudwatch alarms that will be setup along with the ASG. Note the alarm name will be the name attribute defined, plus the ASG resource name.

Ensure myasg exists:
  boto_asg.present:
    - name: myasg
    - launch_config_name: mylc
    - availability_zones:
      - us-east-1a
      - us-east-1b
    - min_size: 1
    - max_size: 1
    - desired_capacity: 1
    - load_balancers:
      - myelb
    - profile: myprofile
    - alarms:
        CPU:
          name: 'ASG CPU **MANAGED BY SALT**'
          attributes:
            metric: CPUUtilization
            namespace: AWS/EC2
            statistic: Average
            comparison: '>='
            threshold: 65.0
            period: 60
            evaluation_periods: 30
            unit: null
            description: 'ASG CPU'
            alarm_actions: [ 'arn:aws:sns:us-east-1:12345:myalarm' ]
            insufficient_data_actions: []
            ok_actions: [ 'arn:aws:sns:us-east-1:12345:myalarm' ]

You can also use alarms from pillars, and override values from the pillar alarms by setting overrides on the resource. Note that 'boto_asg_alarms' will be used as a default value for all resources, if defined and can be used to ensure alarms are always set for an ASG resource.

Setting the alarms in a pillar:

my_asg_alarm:
  CPU:
    name: 'ASG CPU **MANAGED BY SALT**'
    attributes:
      metric: CPUUtilization
      namespace: AWS/EC2
      statistic: Average
      comparison: '>='
      threshold: 65.0
      period: 60
      evaluation_periods: 30
      unit: null
      description: 'ASG CPU'
      alarm_actions: [ 'arn:aws:sns:us-east-1:12345:myalarm' ]
      insufficient_data_actions: []
      ok_actions: [ 'arn:aws:sns:us-east-1:12345:myalarm' ]

Overriding the alarm values on the resource:

Ensure myasg exists:
  boto_asg.present:
    - name: myasg
    - launch_config_name: mylc
    - availability_zones:
      - us-east-1a
      - us-east-1b
    - min_size: 1
    - max_size: 1
    - desired_capacity: 1
    - load_balancers:
      - myelb
    - profile: myprofile
    - alarms_from_pillar: my_asg_alarm
    # override CPU:attributes:threshold
    - alarms:
        CPU:
          attributes:
            threshold: 50.0

salt.states.boto_asg.absent(name, force=False, region=None, key=None, keyid=None, profile=None)¶

Ensure the named autoscale group is deleted.

name: Name of the autoscale group.
force: Force deletion of autoscale group.
region: The region to connect to.
key: Secret key to be used.
keyid: Access key to be used.
profile: A dict with region, key and keyid, or a pillar key (string) that contains a dict with region, key and keyid.

salt.states.boto_asg.present(name, launch_config_name, availability_zones, min_size, max_size, launch_config=None, desired_capacity=None, load_balancers=None, default_cooldown=None, health_check_type=None, health_check_period=None, placement_group=None, vpc_zone_identifier=None, tags=None, termination_policies=None, suspended_processes=None, scaling_policies=None, scaling_policies_from_pillar='boto_asg_scaling_policies', alarms=None, alarms_from_pillar='boto_asg_alarms', region=None, key=None, keyid=None, profile=None, notification_arn=None, notification_arn_from_pillar='boto_asg_notification_arn', notification_types=None, notification_types_from_pillar='boto_asg_notification_types')¶

Ensure the autoscale group exists.

name: Name of the autoscale group.

launch_config_name Name of the launch config to use for the group. Or, if launch_config is specified, this will be the launch config name's prefix. (see below)

launch_config A dictionary of launch config attributes. If specified, a launch config will be used or created, matching this set of attributes, and the autoscale group will be set to use that launch config. The launch config name will be the launch_config_name followed by a hyphen followed by a hash of the launch_config dict contents.

availability_zones

List of availability zones for the group.

min_size

Minimum size of the group.

max_size

Maximum size of the group.

desired_capacity

The desired capacity of the group.

load_balancers

List of load balancers for the group. Once set this can not be updated (Amazon restriction).

default_cooldown

Number of seconds after a Scaling Activity completes before any further scaling activities can start.

health_check_type

The service you want the health status from, Amazon EC2 or Elastic Load Balancer (EC2 or ELB).

health_check_period

Length of time in seconds after a new EC2 instance comes into service that Auto Scaling starts checking its health.

placement_group

Physical location of your cluster placement group created in Amazon EC2. Once set this can not be updated (Amazon restriction).

vpc_zone_identifier

A list of the subnet identifiers of the Virtual Private Cloud.

tags

A list of tags. Example:

key: 'key' value: 'value' propagate_at_launch: true

termination_policies

A list of termination policies. Valid values are: “OldestInstance”, “NewestInstance”, “OldestLaunchConfiguration”, “ClosestToNextInstanceHour”, “Default”. If no value is specified, the “Default” value is used.

suspended_processes

List of processes to be suspended. see http://docs.aws.amazon.com/AutoScaling/latest/DeveloperGuide/US_SuspendResume.html

scaling_policies

List of scaling policies. Each policy is a dict of key-values described by http://boto.readthedocs.org/en/latest/ref/autoscale.html#boto.ec2.autoscale.policy.ScalingPolicy

scaling_policies_from_pillar:

name of pillar dict that contains scaling policy settings. Scaling policies defined for this specific state will override those from pillar.

alarms:

a dictionary of name->boto_cloudwatch_alarm sections to be associated with this ASG. All attributes should be specified except for dimension which will be automatically set to this ASG. See the boto_cloudwatch_alarm state for information about these attributes.

alarms_from_pillar:

name of pillar dict that contains alarm settings. Alarms defined for this specific state will override those from pillar.

region

The region to connect to.

key

Secret key to be used.

keyid

Access key to be used.

profile

A dict with region, key and keyid, or a pillar key (string) that contains a dict with region, key and keyid.

notification_arn

The aws arn that notifications will be sent to

notification_arn_from_pillar

name of the pillar dict that contains notifcation_arn settings. A notification_arn defined for this specific state will override the one from pillar.

notification_types

A list of event names that will trigger a notification. The list of valid notification types is:

"autoscaling:EC2_INSTANCE_LAUNCH", "autoscaling:EC2_INSTANCE_LAUNCH_ERROR", "autoscaling:EC2_INSTANCE_TERMINATE", "autoscaling:EC2_INSTANCE_TERMINATE_ERROR", "autoscaling:TEST_NOTIFICATION"

notification_types_from_pillar

name of the pillar dict that contains notifcation_types settings. Notification_types defined for this specific state will override those from the pillar.