Make ec2_instance able to create new instance or fail if instance matching filter already found

[Elias481]

SUMMARY

It would be fine if the filter in ec2_instance could be used to create new instance if no instance found matching filter criteria and fail otherwise without touching an already running instance.

ISSUE TYPE

• Feature Idea

COMPONENT NAME

ec2_instance

ADDITIONAL INFORMATION

[ansibot]

Files identified in the description:

• lib/ansible/modules/cloud/amazon/ec2_instance.py

If these files are inaccurate, please update the component name section of the description or use the !component bot command.

click here for bot help

[ansibot]

cc @Shaps @ryansb
click here for bot help

[ansibot]

@Elias481, just so you are aware we have a dedicated Working Group for aws.
You can find other people interested in this in #ansible-aws on Freenode IRC
For more information about communities, meetings and agendas see https://github.com/ansible/community

click here for bot help

[Shaps]

@Elias481, thanks for submitting this, I don't think this should be included in the module itself, the module ( and ansible in general ) is to ensure you have the resource/s in the desired state; That said, if this is a feature you need, you can already achieve it by using ec2_instance_facts/failed_when:

- name: Fail if instance already exist
  ec2_instance_facts:
    [...]
    filters:
      <your_filters>
  register: instance_exists
  failed_when: instance_exists.instances | length > 0

[Elias481]

Yes for sure it's possible to check the in yet another additional ansible block.
Its just that I think the use-case is quite common. For setup of new servers an instance is created, according to specs.
During lifetime it's most often not the case that adjuments to the instance would be done with involving ansible and also it's difficult to ensure that the changes are always populated to ansible config.

It is always the case and the approach is symply to say users (acting as admins) that all changes that they do to the system without commiting it to ansible will be overwritten.
While this can be quite safely handled or at least leads to only partly rolled-back changes with (hopefully) learning effect, it's very dangerous to apply changed settings to an instance by mistake with next ansible run because that is likely to cause an uncoordinated reboot of the instance.

For me it's a legitimate definition of having a desired state "instance running" that I just want to ensure the instance is available before continuing the play with the parts that follow which need the instance to be running and having the definition of the details only applied if instance does not exists.

So I do not really want to fail, but just an option that tells the module to not change anything if the instance is already there and populate the results with the found instance.

Of cause an instance_facts step before can be used. But it's not only quite much overhead of ansible script lines (which is anyway very much to implement sane workflows) but also avoids a call to gather-instance-facts for new instances (as the ec2_instance module is querying the same again).

[resmo]

Looking at other cloud modules (vultr, cloustack, cloudscale) there is a logic to not changing a running instance if the change would require a reboot unless force=true is given.

https://docs.ansible.com/ansible/latest/modules/vultr_server_module.html#parameters

[Shaps]

@resmo sensible comment, I like consistency, so I'm happy to add the force parameter to the module. That said, I'd set it to true by default to not drastically change the module's behavior. Thoughts?

[resmo]

@Shaps for "backwards compatibility" it make sense to set it to "true".

[ansibot]

cc @jillr @s-hertel @tremble
click here for bot help

[ansibot]

Thank you very much for your interest in Ansible. Ansible has migrated much of the content into separate repositories to allow for more rapid, independent development. We are closing this issue/PR because this content has been moved to one or more collection repositories.

• lib/ansible/modules/cloud/amazon/ec2_instance.py -> https://galaxy.ansible.com/community/aws

For further information, please see:
https://github.com/ansible/ansibullbot/blob/master/docs/collection_migration.md