NAS-130905 / 25.04 / Remove REST from test_050_alert.py (#14415)

* remove numbers from test and module names

* remove rest and simplify

* replace global variables with pytest cache

* address @yocalebo

* change quotes

* fix `get_alert_by_id` and account for cache dependency

* use `alert.run_source`

* use fixtures

* fix indexing issues

* can't mark a fixture

* fixture timeout

* allow parametrization

* process alerts before waiting for alert to disappear

* Revert "allow parametrization"

This reverts commit 8b7ffe0.

* remove `autouse`

tests/api2/test_alert.py

@@ -0,0 +1,74 @@
+from time import sleep
+
+import pytest
+
+from auto_config import pool_name
+from middlewared.test.integration.utils import call, ssh
+
+
+ID_PATH = "/dev/disk/by-partuuid/"
+
+
+def get_alert_by_id(alert_id):
+    return next(filter(lambda alert: alert["id"] == alert_id, call("alert.list")), None)
+
+
+def wait_for_alert(timeout=120):
+    for _ in range(timeout):
+        for alert in call("alert.list"):
+            if (
+                alert["source"] == "VolumeStatus" and
+                alert["args"]["volume"] == pool_name and
+                alert["args"]["state"] == "DEGRADED"
+            ):
+                return alert["id"]
+        sleep(1)
+
+
+@pytest.fixture(scope="module")
+def degraded_pool_gptid():
+    get_pool = call("pool.query", [["name", "=", pool_name]], {"get": True})
+    gptid = get_pool["topology"]["data"][0]["path"].replace(ID_PATH, "")
+    ssh(f"zinject -d {gptid} -A fault {pool_name}")
+    return gptid
+
+
+@pytest.fixture(scope="module")
+def alert_id(degraded_pool_gptid):
+    call("alert.process_alerts")
+    result = wait_for_alert()
+    if result is None:
+        pytest.fail("Timed out while waiting for alert.")
+    return result
+
+
+def test_verify_the_pool_is_degraded(degraded_pool_gptid):
+    status = call("zpool.status", {"name": pool_name})
+    disk_status = status[pool_name]["data"][ID_PATH + degraded_pool_gptid]["disk_status"]
+    assert disk_status == "DEGRADED"
+
+
+def test_dismiss_alert(alert_id):
+    call("alert.dismiss", alert_id)
+    alert = get_alert_by_id(alert_id)
+    assert alert["dismissed"] is True, alert
+
+
+def test_restore_alert(alert_id):
+    call("alert.restore", alert_id)
+    alert = get_alert_by_id(alert_id)
+    assert alert["dismissed"] is False, alert
+
+
+def test_clear_the_pool_degradation(degraded_pool_gptid):
+    ssh(f"zpool clear {pool_name}")
+    status = call("zpool.status", {"name": pool_name})
+    disk_status = status[pool_name]["data"][ID_PATH + degraded_pool_gptid]["disk_status"]
+    assert disk_status != "DEGRADED"
+
+
+@pytest.mark.timeout(120)
+def test_wait_for_the_alert_to_disappear(alert_id):
+    call("alert.process_alerts")
+    while get_alert_by_id(alert_id) is not None: 
+        sleep(1)