There are a number of ways to handle fatal errors.

• Use fail to end a play and return "failed" 
• Use the --check command line flag to run the playbook in a dry run mode
• Use check_mode to determine if any fatal errors would be returned
• Use failed_when to continue based on the task return code
• Use ignore_errors to ignore errors returned by a task
• Use meta: clear_host_errors (this article)
• Use meta: end_play to end a play for all hosts
• Use meta: end_host to end a play for certain hosts and return "skipped"
• Use block rescue always to run tasks even after a fatal error is returned
• Use stat to determine if /tmp/foo.txt exists and the when parameter to skip the lineinfile module if /tmp/foo.txt does not exist.
• Use stat to determine if /tmp/foo.txt exists and then use the file module and the when parameter to create /tmp/foo.txt if the file does not exist

The meta module with clear_host_errors can be used to clear fatal and fail errors.

The ansible-doc meta command can be used to show the Ansible documention on the meta module.

Let's say /tmp/foo.txt does not exist and you attempt to append "Hello World" to /tmp/foo.txt using the lineinfile module.

---
- hosts: all
  tasks:
  - name: append 'Hello World' to /tmp/foo.txt
    lineinfile:
      path: /tmp/foo.txt
      line: Hello World
...

When the target file does not exist, invoking this playbook will return a fatal error, like this.

TASK [append 'Hello World' to /tmp/foo.txt] 
fatal: [server1.example.com]: FAILED! => {"ansible_facts": {"discovered_interpreter_python": "/usr/bin/python"}, "changed": false, "msg": "Destination /tmp/foo.txt does not exist !", "rc": 257}

The meta module with clear_host_errors will change this behavior, so that server1.example.com is processed in the subsequent tasks.

---
- hosts: all
  tasks:
  - name: append 'Hello World' to /tmp/foo.txt
    lineinfile:
      path: /tmp/foo.txt
      line: Hello World

  - meta: clear_host_errors
...