I am creating a Windows instance using Ansible. Once the instance got created I need to install IIS on that created instance automatically.
How to login to the newly created instance automatically?
What are the parameters I need to mention in the host (inventory) file to login?
To connect to ec2 instances you will first need to retrieve the password using the ec2_win_password module
Example:
- ec2_win_password:
instance_id: i-xxxxxxx
region: us-west-2
key_file: "/.ssh/yourPEM.pem"
aws_access_key: "{{ aws_access_key }}"
aws_secret_key: "{{ aws_secret_key}}"
security_token: "{{ session_token }}"
key_passphrase: "1122355"
register: result
- debug : msg="{{ result }}"
The result can be used as your ansible_password
Related
I have a basic Ansible playbook creating an ec2 instance in aws, its working fine, i want to pick the public ip from the created instance to do furthers task on it, i am using the following section:
- name: get-facts
ec2_instance_facts:
region: eu-west-1
aws_access_key: xxxxxxxxx
aws_secret_key: xxxxxxx
filters:
"tag:Name": docker-server
register: ec2
- debug: var=ec2.instances
But i want to get just the ip address, according to the documentation public_ip
is a string, but i cannot call it when i use ec2.instances.public_ip i get error:
"VARIABLE IS NOT DEFINED!"
what can i do to get the public ip for my created ec2 instance ? thanks.
It is ec2.instances.public_ip_address and public_ip
See: https://docs.ansible.com/ansible/2.9/modules/ec2_instance_info_module.html#return-instances/public_ip_address
ec2_instance_facts deprecated, ansible offer to use ec2_instance_info
See: https://docs.ansible.com/ansible/2.9/modules/ec2_instance_facts_module.html
UPDATE:
- name: Print info
debug: var="ec2.instances[0].public_ip_address"
- name: Loop instances
debug:
var: item.public_ip_address
with_items: "{{ ec2.instances }}"
Thank you for your support and clarification, its finally worked like this:
- name: get-facts
ec2_instance_info:
region: xxxxx
aws_access_key: xxxxxxxxxxx
aws_secret_key: xxxxxxxxxxx
filters:
"tag:Name": docker-server
register: ec2
- debug: var="ec2.instances[0].public_ip_address"
I am launching aws ec2 2 instances using ansible using count:2 please check below playbook
- name: Create an EC2 instance
ec2:
aws_access_key: "{{ access_key }}"
aws_secret_key: "{{ key }}"
key_name: "{{ keypair }}"
region: "{{ region }}"
group: "{{ security_group }}"
instance_type: "{{ instance_type }}"
image: "{{ image }}"
wait: yes
count: 2
vpc_subnet_id: "{{ vpc_subnet_id }}"
assign_public_ip: "{{ assign_public_ip }}"
register: ec2
- name: Add the newly created 1 EC2 instance(s) to webserver group
lineinfile: dest=inventory
insertafter='^\[webserver\]$'
line="{{ item.private_ip }} {{hoststring}}"
state=present
with_items: "{{ ec2.instances }}"
- name: add newly created remaining ec2 instance to db group
lineinfile: dest=inventory
insertafter='^\[db-server\]$'
line="{{ item.private_ip }} {{hoststring}}"
state=present
with_items: "{{ ec2.instances }}"
Here i want to add one ip to webserver host group & remaining to db host group but its not working with above playbook please help me to achieve same?
i dont wnt to use add_host here.
Since you are using AWS, have you considered using the aws_ec2 plugin for dynamic inventory?
As long as you are tagging your instances correctly, and you set up the yaml file, it would do what you want it.
Otherwise, your register: ec2 has two elements in it. The way (if it worked) you are looping through ec2 would add both to each group. You would need to add a when condition to match the something like the tag/subnet/cidr to know which server to add to which group.
One way to help see what the return is would be do print out the ec2 variable:
- debug: var=ec2
I'm trying to use Ansible to create two instances, each instance must have different subnet ids. I'm using exact_count with tag Name to keep track of instances. The main issue is that I am not able to understand how to provide two different subnet ids in the same playbook.
This task can be used in order to create the multiple ec2 instances with different subnets
- name: 7. Create EC2 server
ec2:
image: "{{ image }}"
wait: true
instance_type: t2.micro
group_id: "{{ security_group.group_id }}"
vpc_subnet_id: "{{ item }}"
key_name: "{{ key_name }}"
count: 1
region: us-east-1
with_items:
- "{{ subnet1.subnet.id }}"
- "{{ subnet2.subnet.id }}"
register: ec2
For ELB, if we want to remove the instance from all the elbs, we just need to pass the instance id to the elb_instance module and again if we want to add the instance back in the same playbook, it give us a magic variable ec2_elbs, we can iterate over this variable and add back or register the instance to all the elbs which he was registered previously.
I didn't find any module, which I can use to find the list of the target groups in which the instance is register. can somebody point if know about it.
Here is elb_target_group module which is in preview mode or you can make use of a python script and boto to find the list of instances in a particular targetgroup
I have found the way to dynamically removed the instance(s) from all the target groups that it has been registered:
- name: Collect facts about EC2 Instance
action: ec2_metadata_facts
- name: Get the list of all target groups to which Instance(s) registered
local_action:
module: elb_target_facts
instance_id: "{{ ansible_ec2_instance_id }}"
region: "{{ ansible_ec2_placement_region }}"
register: alb_target_facts
become: no
- name: Deregister the Instance from all target groups
local_action:
module: elb_target
target_group_arn: "{{ item.target_group_arn }}"
target_id: "{{ ansible_ec2_instance_id }}"
region: "{{ ansible_ec2_placement_region }}"
target_status_timeout: 300
target_status: unused
state: absent
loop: "{{ alb_target_facts.instance_target_groups }}"
become: no
Because I have registered the target group(s) information in a variable, I can use it again to add it back:
- name: Register the Instance back to the target group(s)
local_action:
module: elb_target
target_group_arn: "{{ item.target_group_arn }}"
target_id: "{{ ansible_ec2_instance_id }}"
region: "{{ ansible_ec2_placement_region }}"
target_status: healthy
target_status_timeout: 300
state: present
loop: "{{ alb_target_facts.instance_target_groups }}"
become: no
I am trying to modify an AWS instance from within ansible.
My playbook is designed to compile and install ixgbevf driver on a running AWS instance then shuts it down so that a ec2-modify-instance-attribute command can be run.
Everything up to the last part of the equation works.
My final role stanza looks like this:
- name: Turn on enhanced networking on on the AMI host
local_action: ec2-modify-instance-attribute the_instance_id={{ my_instance_id }} --sriov simple --region us-west-2
with_items: ec2_info.instances
ignore_errors: yes
When the role runs I get the following errror:
fatal: [10.11.31.48 -> 127.0.0.1] => module ec2-modify-instance-attribute not found in configured module paths
FATAL: all hosts have already failed -- aborting
Is there actually a way to use the ec2-modify-instance-attribute command from within Ansible?
Try using this CLI workaround
local_action:
module: >
shell
aws ec2 modify-instance-attribute --instance-id "{{ ec2_id }}"
--groups "{{ec2_security_group_ids}}" "{{newgroup.group_id}}"
when: newgroup.group_id not in ec2_security_group_ids
I ended up with the following which worked:
- name: "Stop the AWS instance"
local_action:
module: ec2
state: stopped
aws_access_key: "{{ aws.ec2.access_key_id }}"
aws_secret_key: "{{ aws.ec2.secret_access_key }}"
instance_ids: "{{ ansible_ec2_instance_id }}"
region: "{{ ansible_ec2_placement_region }}"
wait: True