1
0
Fork 0
mirror of https://github.com/ansible-collections/community.general.git synced 2024-09-14 20:13:21 +02:00
community.general/test/integration/targets/ec2_asg/tasks/main.yml
msven a2b3120e85 ec2_asg: fix #28087 and #35993 (#36679)
Fixes #35993 - Changes to update_size in commit eb4cc31 made it so
the group dict passed into update_size was not modified. As a result,
the 'replace' call does not see an updated min_size like it previously
did and doesn't pause to wait for any new instances to spin up. Instead,
it moves straight into terminating old instances. Fix is to add batch_size
to min_size when calling wait_for_new_inst.

Fixes #28087 - Make replace_all_instances and replace_instances behave
exactly the same by setting replace_instances = current list of instances
when replace_all_instances used. Root cause of issue was that without lc_check
terminate_batch will terminate all instances passed to it and after updating
the asg size we were querying the asg again for the list of instances - so terminate batch
saw the list including new ones just spun up.

When creating new asg with replace_all_instances: yes and lc_check: false
the instances that are initially created are then subsequently replaced.
This change makes it so replace only occurs if the asg already existed.

Add integration tests for #28087 and #35993.
2018-03-05 11:47:31 -05:00

619 lines
18 KiB
YAML

---
# tasks file for test_ec2_asg
- name: Test incomplete credentials with ec2_asg
block:
# ============================================================
- name: test invalid profile
ec2_asg:
name: "{{ resource_prefix }}-asg"
region: "{{ aws_region }}"
profile: notavalidprofile
ignore_errors: yes
register: result
- name:
assert:
that:
- "'The config profile (notavalidprofile) could not be found' in result.msg"
- name: test partial credentials
ec2_asg:
name: "{{ resource_prefix }}-asg"
region: "{{ aws_region }}"
aws_access_key: "{{ aws_access_key }}"
ignore_errors: yes
register: result
- name:
assert:
that:
- "'Partial credentials found in explicit, missing: aws_secret_access_key' in result.msg"
- name: test without specifying region
ec2_asg:
name: "{{ resource_prefix }}-asg"
aws_access_key: "{{ aws_access_key }}"
aws_secret_key: "{{ aws_secret_key }}"
security_token: "{{ security_token }}"
ignore_errors: yes
register: result
- name:
assert:
that:
- result.msg == 'The ec2_asg module requires a region and none was found in configuration, environment variables or module parameters'
# ============================================================
- name: Test incomplete arguments with ec2_asg
block:
# ============================================================
- name: test without specifying required module options
ec2_asg:
aws_access_key: "{{ aws_access_key }}"
aws_secret_key: "{{ aws_secret_key }}"
security_token: "{{ security_token }}"
ignore_errors: yes
register: result
- name: assert name is a required module option
assert:
that:
- "result.msg == 'missing required arguments: name'"
- name: Run ec2_asg integration tests.
block:
# ============================================================
# Set up the testing dependencies: VPC, subnet, security group, and two launch configurations
- name: set connection information for all tasks
set_fact:
aws_connection_info: &aws_connection_info
aws_access_key: "{{ aws_access_key }}"
aws_secret_key: "{{ aws_secret_key }}"
security_token: "{{ security_token }}"
region: "{{ aws_region }}"
no_log: yes
- name: Create VPC for use in testing
ec2_vpc_net:
name: "{{ resource_prefix }}-vpc"
cidr_block: 10.55.77.0/24
tenancy: default
<<: *aws_connection_info
register: testing_vpc
- name: Create internet gateway for use in testing
ec2_vpc_igw:
vpc_id: "{{ testing_vpc.vpc.id }}"
state: present
<<: *aws_connection_info
register: igw
- name: Create subnet for use in testing
ec2_vpc_subnet:
state: present
vpc_id: "{{ testing_vpc.vpc.id }}"
cidr: 10.55.77.0/24
az: "{{ aws_region }}a"
resource_tags:
Name: "{{ resource_prefix }}-subnet"
<<: *aws_connection_info
register: testing_subnet
- name: create routing rules
ec2_vpc_route_table:
vpc_id: "{{ testing_vpc.vpc.id }}"
tags:
created: "{{ resource_prefix }}-route"
routes:
- dest: 0.0.0.0/0
gateway_id: "{{ igw.gateway_id }}"
subnets:
- "{{ testing_subnet.subnet.id }}"
<<: *aws_connection_info
- name: create a security group with the vpc created in the ec2_setup
ec2_group:
name: "{{ resource_prefix }}-sg"
description: a security group for ansible tests
vpc_id: "{{ testing_vpc.vpc.id }}"
rules:
- proto: tcp
from_port: 22
to_port: 22
cidr_ip: 0.0.0.0/0
- proto: tcp
from_port: 80
to_port: 80
cidr_ip: 0.0.0.0/0
<<: *aws_connection_info
register: sg
- name: ensure launch configs exist
ec2_lc:
name: "{{ item }}"
assign_public_ip: true
image_id: "{{ ec2_ami_image[aws_region] }}"
user_data: |
#cloud-config
package_upgrade: true
package_update: true
packages:
- httpd
runcmd:
- "service httpd start"
security_groups: "{{ sg.group_id }}"
instance_type: t2.micro
<<: *aws_connection_info
with_items:
- "{{ resource_prefix }}-lc"
- "{{ resource_prefix }}-lc-2"
# ============================================================
- name: launch asg and wait for instances to be deemed healthy (no ELB)
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
desired_capacity: 1
min_size: 1
max_size: 1
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
state: present
wait_for_instances: yes
<<: *aws_connection_info
register: output
- assert:
that:
- "output.viable_instances == 1"
# - name: pause for a bit to make sure that the group can't be trivially deleted
# pause: seconds=30
- name: kill asg
ec2_asg:
name: "{{ resource_prefix }}-asg"
state: absent
wait_timeout: 700
<<: *aws_connection_info
async: 300
# ============================================================
- name: launch asg and do not wait for instances to be deemed healthy (no ELB)
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
desired_capacity: 1
min_size: 1
max_size: 1
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
wait_for_instances: no
state: present
<<: *aws_connection_info
register: output
- assert:
that:
- "output.viable_instances == 0"
- name: kill asg
ec2_asg:
name: "{{ resource_prefix }}-asg"
state: absent
<<: *aws_connection_info
async: 300
# ============================================================
- name: create asg with asg metrics enabled
ec2_asg:
name: "{{ resource_prefix }}-asg"
metrics_collection: true
launch_config_name: "{{ resource_prefix }}-lc"
desired_capacity: 0
min_size: 0
max_size: 0
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
state: present
<<: *aws_connection_info
register: output
- assert:
that:
- "'Group' in output.metrics_collection.0.Metric"
- name: kill asg
ec2_asg:
name: "{{ resource_prefix }}-asg"
state: absent
<<: *aws_connection_info
async: 300
# ============================================================
- name: load balancer name has to be less than 32 characters
# the 8 digit identifier at the end of resource_prefix helps determine during which test something
# was created
set_fact:
load_balancer_name: "{{ item }}-lb"
with_items: "{{ resource_prefix | regex_findall('.{8}$') }}"
- name: launch load balancer
ec2_elb_lb:
name: "{{ load_balancer_name }}"
state: present
security_group_ids:
- "{{ sg.group_id }}"
subnets: "{{ testing_subnet.subnet.id }}"
connection_draining_timeout: 60
listeners:
- protocol: http
load_balancer_port: 80
instance_port: 80
health_check:
ping_protocol: tcp
ping_port: 80
ping_path: "/"
response_timeout: 5
interval: 10
unhealthy_threshold: 4
healthy_threshold: 2
<<: *aws_connection_info
register: load_balancer
- name: launch asg and wait for instances to be deemed healthy (ELB)
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
health_check_type: ELB
desired_capacity: 1
min_size: 1
max_size: 1
health_check_period: 300
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
load_balancers: "{{ load_balancer_name }}"
wait_for_instances: yes
wait_timeout: 900
state: present
<<: *aws_connection_info
register: output
- assert:
that:
- "output.viable_instances == 1"
# ============================================================
# grow scaling group to 3
- name: add 2 more instances wait for instances to be deemed healthy (ELB)
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
health_check_type: ELB
desired_capacity: 3
min_size: 3
max_size: 5
health_check_period: 600
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
load_balancers: "{{ load_balancer_name }}"
wait_for_instances: yes
wait_timeout: 1200
state: present
<<: *aws_connection_info
register: output
- assert:
that:
- "output.viable_instances == 3"
# ============================================================
# # perform rolling replace with different launch configuration
- name: perform rolling update to new AMI
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc-2"
health_check_type: ELB
desired_capacity: 3
min_size: 1
max_size: 5
health_check_period: 900
load_balancers: "{{ load_balancer_name }}"
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
wait_for_instances: yes
replace_all_instances: yes
wait_timeout: 1800
state: present
<<: *aws_connection_info
register: output
# ensure that all instances have new launch config
- assert:
that:
- "item.value.launch_config_name == '{{ resource_prefix }}-lc-2'"
with_dict: "{{ output.instance_facts }}"
# assert they are all healthy and that the rolling update resulted in the appropriate number of instances
- assert:
that:
- "output.viable_instances == 3"
# ============================================================
# perform rolling replace with the original launch configuration
- name: perform rolling update to new AMI while removing the load balancer
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
health_check_type: EC2
desired_capacity: 3
min_size: 1
max_size: 5
health_check_period: 900
load_balancers: []
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
wait_for_instances: yes
replace_all_instances: yes
wait_timeout: 1800
state: present
<<: *aws_connection_info
register: output
# ensure that all instances have new launch config
- assert:
that:
- "item.value.launch_config_name == '{{ resource_prefix }}-lc'"
with_dict: "{{ output.instance_facts }}"
# assert they are all healthy and that the rolling update resulted in the appropriate number of instances
# there should be the same number of instances as there were before the rolling update was performed
- assert:
that:
- "output.viable_instances == 3"
# ============================================================
# perform rolling replace with new launch configuration and lc_check:false
# Note - this is done async so we can query asg_facts during
# the execution. Issues #28087 and #35993 result in correct
# end result, but spin up extraneous instances during execution.
- name: "perform rolling update to new AMI with lc_check: false"
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc-2"
health_check_type: EC2
desired_capacity: 3
min_size: 1
max_size: 5
health_check_period: 900
load_balancers: []
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
wait_for_instances: yes
replace_all_instances: yes
replace_batch_size: 3
lc_check: false
wait_timeout: 1800
state: present
<<: *aws_connection_info
async: 1800
poll: 0
register: asg_job
- name: get ec2_asg facts for 3 minutes
ec2_asg_facts:
name: "{{ resource_prefix }}-asg"
<<: *aws_connection_info
register: output
loop_control:
pause: 15
with_sequence: count=12
- set_fact:
inst_id_json_query: 'results[*].results[*].instances[*].instance_id'
# Since we started with 3 servers and replace all of them.
# We should see 6 servers total.
- assert:
that:
- "lookup('flattened',output|json_query(inst_id_json_query)).split(',')|unique|length == 6"
- name: Ensure ec2_asg task completes
async_status: jid="{{ asg_job.ansible_job_id }}"
register: status
until: status.finished
retries: 200
delay: 15
# ============================================================
- name: kill asg
ec2_asg:
name: "{{ resource_prefix }}-asg"
state: absent
<<: *aws_connection_info
async: 300
# Create new asg with replace_all_instances and lc_check:false
# Note - this is done async so we can query asg_facts during
# the execution. Issues #28087 results in correct
# end result, but spin up extraneous instances during execution.
- name: "new asg with lc_check: false"
ec2_asg:
name: "{{ resource_prefix }}-asg"
launch_config_name: "{{ resource_prefix }}-lc"
health_check_type: EC2
desired_capacity: 3
min_size: 1
max_size: 5
health_check_period: 900
load_balancers: []
vpc_zone_identifier: "{{ testing_subnet.subnet.id }}"
wait_for_instances: yes
replace_all_instances: yes
replace_batch_size: 3
lc_check: false
wait_timeout: 1800
state: present
<<: *aws_connection_info
async: 1800
poll: 0
register: asg_job
# Collect ec2_asg_facts for 3 minutes
- name: get ec2_asg facts
ec2_asg_facts:
name: "{{ resource_prefix }}-asg"
<<: *aws_connection_info
register: output
loop_control:
pause: 15
with_sequence: count=12
- set_fact:
inst_id_json_query: 'results[*].results[*].instances[*].instance_id'
# Get all instance_ids we saw and assert we saw number expected
# Should only see 3 (don't replace instances we just created)
- assert:
that:
- "lookup('flattened',output|json_query(inst_id_json_query)).split(',')|unique|length == 3"
- name: Ensure ec2_asg task completes
async_status: jid="{{ asg_job.ansible_job_id }}"
register: status
until: status.finished
retries: 200
delay: 15
# ============================================================
always:
- name: kill asg
ec2_asg:
name: "{{ resource_prefix }}-asg"
state: absent
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
# Remove the testing dependencies
- name: remove the load balancer
ec2_elb_lb:
name: "{{ load_balancer_name }}"
state: absent
security_group_ids:
- "{{ sg.group_id }}"
subnets: "{{ testing_subnet.subnet.id }}"
wait: yes
connection_draining_timeout: 60
listeners:
- protocol: http
load_balancer_port: 80
instance_port: 80
health_check:
ping_protocol: tcp
ping_port: 80
ping_path: "/"
response_timeout: 5
interval: 10
unhealthy_threshold: 4
healthy_threshold: 2
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
- name: remove launch configs
ec2_lc:
name: "{{ resource_prefix }}-lc"
state: absent
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
with_items:
- "{{ resource_prefix }}-lc"
- "{{ resource_prefix }}-lc-2"
- name: remove the security group
ec2_group:
name: "{{ resource_prefix }}-sg"
description: a security group for ansible tests
vpc_id: "{{ testing_vpc.vpc.id }}"
state: absent
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
- name: remove routing rules
ec2_vpc_route_table:
state: absent
vpc_id: "{{ testing_vpc.vpc.id }}"
tags:
created: "{{ resource_prefix }}-route"
routes:
- dest: 0.0.0.0/0
gateway_id: "{{ igw.gateway_id }}"
subnets:
- "{{ testing_subnet.subnet.id }}"
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
- name: remove internet gateway
ec2_vpc_igw:
vpc_id: "{{ testing_vpc.vpc.id }}"
state: absent
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
- name: remove the subnet
ec2_vpc_subnet:
state: absent
vpc_id: "{{ testing_vpc.vpc.id }}"
cidr: 10.55.77.0/24
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10
- name: remove the VPC
ec2_vpc_net:
name: "{{ resource_prefix }}-vpc"
cidr_block: 10.55.77.0/24
state: absent
<<: *aws_connection_info
register: removed
until: removed is not failed
ignore_errors: yes
retries: 10