mitogen/tests/ansible/integration/context_service/reconnection.yml

# Test ContextService ability to handle disconnections, including handling
# cleanup of dependent (via=) contexts.

- name: integration/context_service/reconnection.yml
  hosts: test-targets
  any_errors_fatal: true
  tasks:
    - mitogen_shutdown_all:

    - custom_python_detect_environment:
      register: ssh_account_env

    - become: true
      custom_python_detect_environment:
      register: old_become_env

    - become: true
      shell: |
        bash -c "( sleep 3; kill -9 {{ssh_account_env.pid}}; ) & disown"

    - connection: local
      shell: sleep 3

    - wait_for_connection:

    - become: true
      custom_python_detect_environment:
      register: new_become_env

    # Verify the PIDs really changed (i.e. disconnection happened)
    - assert:
        that:
          - old_become_env.pid != new_become_env.pid
        fail_msg: old_become_env={{old_become_env}} new_become_env={{new_become_env}}
issue #202: ansible: forget all dependent contexts on Stream disconnect This is a partial fix, there are still at least 2 cases needing covered: - In-progress connections must have CallError or similar sent to any waiters - Once connection delegation exists, it is possible for other worker processes to be active (and in any step in the process), trying to communicate with a context that we know can no longer be communicated with. The solution to that isn't clear yet. Additionally ensure root has /bin/bash shell in both Docker images. 7 years ago			`# Test ContextService ability to handle disconnections, including handling`
			`# cleanup of dependent (via=) contexts.`

			`- name: integration/context_service/reconnection.yml`
tests: use test-targets group, not all group 7 years ago			`hosts: test-targets`
issue #202: ansible: forget all dependent contexts on Stream disconnect This is a partial fix, there are still at least 2 cases needing covered: - In-progress connections must have CallError or similar sent to any waiters - Once connection delegation exists, it is possible for other worker processes to be active (and in any step in the process), trying to communicate with a context that we know can no longer be communicated with. The solution to that isn't clear yet. Additionally ensure root has /bin/bash shell in both Docker images. 7 years ago			`any_errors_fatal: true`
			`tasks:`
issue #369: don't mass-kill all SSH clients in reconnection.yml It breaks my new development environment :) 6 years ago			`- mitogen_shutdown_all:`

			`- custom_python_detect_environment:`
			`register: ssh_account_env`
issue #202: ansible: forget all dependent contexts on Stream disconnect This is a partial fix, there are still at least 2 cases needing covered: - In-progress connections must have CallError or similar sent to any waiters - Once connection delegation exists, it is possible for other worker processes to be active (and in any step in the process), trying to communicate with a context that we know can no longer be communicated with. The solution to that isn't clear yet. Additionally ensure root has /bin/bash shell in both Docker images. 7 years ago
			`- become: true`
			`custom_python_detect_environment:`
			`register: old_become_env`

			`- become: true`
			`shell: \|`
issue #369: don't mass-kill all SSH clients in reconnection.yml It breaks my new development environment :) 6 years ago			`bash -c "( sleep 3; kill -9 {{ssh_account_env.pid}}; ) & disown"`
issue #202: ansible: forget all dependent contexts on Stream disconnect This is a partial fix, there are still at least 2 cases needing covered: - In-progress connections must have CallError or similar sent to any waiters - Once connection delegation exists, it is possible for other worker processes to be active (and in any step in the process), trying to communicate with a context that we know can no longer be communicated with. The solution to that isn't clear yet. Additionally ensure root has /bin/bash shell in both Docker images. 7 years ago
			`- connection: local`
			`shell: sleep 3`

			`- wait_for_connection:`

			`- become: true`
			`custom_python_detect_environment:`
			`register: new_become_env`

			`# Verify the PIDs really changed (i.e. disconnection happened)`
			`- assert:`
			`that:`
			`- old_become_env.pid != new_become_env.pid`
tests: Print variables on failure of assert tasks Fixes #852 3 years ago			`fail_msg: old_become_env={{old_become_env}} new_become_env={{new_become_env}}`