set apart host backups of other nodes and of former jobs

- Each host backup carries the node it belongs to, read from its source host or its name (`hostcfg-<node>`, a job of this node or its PBS backup ID); both backup lists also return the node name. - The backup list offers This node / All nodes when some backup belongs to another node, with This node as the default, a count that follows the choice and a note with the backups not shown. - Backups whose node cannot be read from their name are listed under All nodes. - A local backup made on this node by a job that no longer exists is marked as Former job, read from its metadata file; backups without it, manual ones and those of other nodes are not marked. - New texts in English and Spanish, modelled in the translation steady-state test. - Tests for the node and the former job read from each backup, and for the cluster node membership notifications.
2026-11-02 10:52:22 +00:00 · 2026-10-14 18:04:42 +02:00
parent f2c5ab1636
commit c9479c6ac1
7 changed files with 358 additions and 7 deletions
@@ -0,0 +1,106 @@
+"""Corosync reports a node that leaves or comes back with its number, in a
+membership line that never says "node". The surviving nodes notify it by the
+name the node has in corosync.conf."""
+import sys
+import tempfile
+import unittest
+from pathlib import Path
+from queue import Empty, Queue
+from unittest.mock import patch
+
+SCRIPTS = Path(__file__).resolve().parents[1]
+sys.path.insert(0, str(SCRIPTS))
+import notification_events  # noqa: E402
+
+CONF = """totem {
+  cluster_name: lab
+}
+nodelist {
+  node {
+    name: pve1
+    nodeid: 1
+    ring0_addr: 10.0.0.1
+  }
+  node {
+    name: pve2
+    nodeid: 2
+    ring0_addr: 10.0.0.2
+  }
+  node {
+    name: pve3
+    nodeid: 3
+    ring0_addr: 10.0.0.3
+  }
+}
+"""
+
+
+class ClusterMembershipTests(unittest.TestCase):
+    def setUp(self):
+        folder = tempfile.TemporaryDirectory()
+        self.addCleanup(folder.cleanup)
+        conf = Path(folder.name) / "corosync.conf"
+        conf.write_text(CONF)
+        real_open = open
+
+        def corosync_conf(path, *args, **kwargs):
+            if str(path) in ("/etc/pve/corosync.conf", "/etc/corosync/corosync.conf"):
+                path = conf
+            return real_open(path, *args, **kwargs)
+
+        self.booting = False
+        for item in (patch("builtins.open", corosync_conf),
+                     patch.object(notification_events.socket, "gethostname", return_value="pve1"),
+                     patch.object(notification_events.startup_grace, "is_real_system_boot",
+                                  side_effect=lambda: self.booting),
+                     patch.object(notification_events._shared_state, "is_startup_health_grace",
+                                  side_effect=lambda: self.booting)):
+            item.start()
+            self.addCleanup(item.stop)
+        self.queue = Queue()
+        self.watcher = notification_events.JournalWatcher(self.queue)
+
+    def notified(self, line):
+        self.watcher._check_cluster_events(line, "corosync")
+        events = []
+        while True:
+            try:
+                event = self.queue.get_nowait()
+            except Empty:
+                return events
+            events.append((event.event_type, event.data.get("node_name")))
+
+    def test_a_node_that_leaves_is_named(self):
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.55) was formed. Members left: 2"),
+                         [("node_disconnect", "pve2")])
+
+    def test_a_node_that_comes_back_is_named(self):
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.5a) was formed. Members joined: 2"),
+                         [("node_reconnect", "pve2")])
+
+    def test_every_node_of_the_line_is_notified(self):
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.60) was formed. Members left: 2 3"),
+                         [("node_disconnect", "pve2"), ("node_disconnect", "pve3")])
+
+    def test_this_node_is_not_notified(self):
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.61) was formed. Members joined: 1"), [])
+
+    def test_while_this_host_boots_only_the_joins_are_left_out(self):
+        self.booting = True
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.62) was formed. Members joined: 2"), [])
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.63) was formed. Members left: 3"),
+                         [("node_disconnect", "pve3")])
+
+    def test_a_node_missing_from_the_configuration_keeps_its_number(self):
+        self.assertEqual(self.notified("[TOTEM ] A new membership (1.64) was formed. Members left: 7"),
+                         [("node_disconnect", "7")])
+
+    def test_other_corosync_lines_notify_nothing(self):
+        for line in ("[CFG   ] Node 2 was shut down by sysadmin", "[QUORUM] Members[2]: 1 3",
+                     "[KNET  ] link: host: 2 link: 0 is down"):
+            with self.subTest(line=line):
+                self.assertEqual(self.notified(line), [])
+
+
+if __name__ == "__main__":
+    unittest.main()