1 files changed, 144 insertions, 0 deletions
diff --git a/src/ansiblelint/rules/ignore_errors.py b/src/ansiblelint/rules/ignore_errors.py
new file mode 100644
index 0000000..4144f2d
--- /dev/null
+++ b/src/ansiblelint/rules/ignore_errors.py
@@ -0,0 +1,144 @@
+"""IgnoreErrorsRule used with ansible-lint."""
+from __future__ import annotations
+
+import sys
+from typing import TYPE_CHECKING
+
+from ansiblelint.rules import AnsibleLintRule
+
+if TYPE_CHECKING:
+    from ansiblelint.file_utils import Lintable
+    from ansiblelint.utils import Task
+
+
+class IgnoreErrorsRule(AnsibleLintRule):
+    """Use failed_when and specify error conditions instead of using ignore_errors."""
+
+    id = "ignore-errors"
+    description = (
+        "Instead of ignoring all errors, ignore the errors only when using ``{{ ansible_check_mode }}``, "
+        "register the errors using ``register``, "
+        "or use ``failed_when:`` and specify acceptable error conditions "
+        "to reduce the risk of ignoring important failures."
+    )
+    severity = "LOW"
+    tags = ["unpredictability"]
+    version_added = "v5.0.7"
+
+    def matchtask(
+        self,
+        task: Task,
+        file: Lintable | None = None,
+    ) -> bool | str:
+        if (
+            task.get("ignore_errors")
+            and task.get("ignore_errors") != "{{ ansible_check_mode }}"
+            and not task.get("register")
+        ):
+            return True
+
+        return False
+
+
+if "pytest" in sys.modules:
+    import pytest
+
+    if TYPE_CHECKING:
+        from ansiblelint.testing import RunFromText  # pylint: disable=ungrouped-imports
+
+    IGNORE_ERRORS_TRUE = """
+- hosts: all
+  tasks:
+    - name: Run apt-get update
+      command: apt-get update
+      ignore_errors: true
+"""
+
+    IGNORE_ERRORS_FALSE = """
+- hosts: all
+  tasks:
+    - name: Run apt-get update
+      command: apt-get update
+      ignore_errors: false
+"""
+
+    IGNORE_ERRORS_CHECK_MODE = """
+- hosts: all
+  tasks:
+    - name: Run apt-get update
+      command: apt-get update
+      ignore_errors: "{{ ansible_check_mode }}"
+"""
+
+    IGNORE_ERRORS_REGISTER = """
+- hosts: all
+  tasks:
+    - name: Run apt-get update
+      command: apt-get update
+      ignore_errors: true
+      register: ignore_errors_register
+"""
+
+    FAILED_WHEN = """
+- hosts: all
+  tasks:
+    - name: Disable apport
+      become: 'yes'
+      lineinfile:
+        line: "enabled=0"
+        dest: /etc/default/apport
+        mode: 0644
+        state: present
+      register: default_apport
+      failed_when: default_apport.rc !=0 and not default_apport.rc == 257
+"""
+
+    @pytest.mark.parametrize(
+        "rule_runner",
+        (IgnoreErrorsRule,),
+        indirect=["rule_runner"],
+    )
+    def test_ignore_errors_true(rule_runner: RunFromText) -> None:
+        """The task uses ignore_errors."""
+        results = rule_runner.run_playbook(IGNORE_ERRORS_TRUE)
+        assert len(results) == 1
+
+    @pytest.mark.parametrize(
+        "rule_runner",
+        (IgnoreErrorsRule,),
+        indirect=["rule_runner"],
+    )
+    def test_ignore_errors_false(rule_runner: RunFromText) -> None:
+        """The task uses ignore_errors: false, oddly enough."""
+        results = rule_runner.run_playbook(IGNORE_ERRORS_FALSE)
+        assert len(results) == 0
+
+    @pytest.mark.parametrize(
+        "rule_runner",
+        (IgnoreErrorsRule,),
+        indirect=["rule_runner"],
+    )
+    def test_ignore_errors_check_mode(rule_runner: RunFromText) -> None:
+        """The task uses ignore_errors: "{{ ansible_check_mode }}"."""
+        results = rule_runner.run_playbook(IGNORE_ERRORS_CHECK_MODE)
+        assert len(results) == 0
+
+    @pytest.mark.parametrize(
+        "rule_runner",
+        (IgnoreErrorsRule,),
+        indirect=["rule_runner"],
+    )
+    def test_ignore_errors_register(rule_runner: RunFromText) -> None:
+        """The task uses ignore_errors: but output is registered and managed."""
+        results = rule_runner.run_playbook(IGNORE_ERRORS_REGISTER)
+        assert len(results) == 0
+
+    @pytest.mark.parametrize(
+        "rule_runner",
+        (IgnoreErrorsRule,),
+        indirect=["rule_runner"],
+    )
+    def test_failed_when(rule_runner: RunFromText) -> None:
+        """Instead of ignore_errors, this task uses failed_when."""
+        results = rule_runner.run_playbook(FAILED_WHEN)
+        assert len(results) == 0