Add option for no-quorum-policy

Adds a config item for what to do when the cluster does not have quorum.
This is useful with stateless services where, e.g., we only need a VIP
and that can be up on a single host with no problem.

Though this would be a good relation data setting, many sites would
prefer to stop the resources rather than have a VIP on multiple hosts,
causing arp issues with the switch.

Closes-bug: #1850829
Change-Id: I961b6b32e7ed23f967b047dd0ecb45b0c0dff49a

config.yaml

@@ -218,3 +218,12 @@ options:
     description: |
       check_crm will generate a critical alert if the failcount of a resource
       has crossed this threshold. Set to 0 or '' to disable.
+  no_quorum_policy:
+    type: string
+    default: "stop"
+    description: |
+      What to do when the cluster does not have quorum. Allowed values:
+        ignore: continue all resource management,
+        freeze: continue resource management, but don’t recover resources from nodes not in the affected partition,
+        stop: stop all resources in the affected cluster partition,
+        suicide: fence all nodes in the affected cluster partition

hooks/hooks.py

@@ -126,6 +126,7 @@ from utils import (
     emit_systemd_overrides_file,
     trigger_corosync_update_from_leader,
     emit_corosync_conf,
+    assess_status_helper,
 )
 
 from charmhelpers.contrib.charmsupport import nrpe
@@ -230,6 +231,15 @@ def config_changed():
             cfg.previous('maintenance-mode') != cfg['maintenance-mode']):
         maintenance_mode(cfg['maintenance-mode'])
 
+    if config('no_quorum_policy').lower() in ['ignore', 'freeze',
+                                              'stop', 'suicide']:
+        pcmk.set_property("no-quorum-policy",
+                          config('no_quorum_policy').lower())
+    else:
+        message = 'Invalid setting for no_quorum_policy'
+        status_set('blocked', message)
+    status_set(*assess_status_helper())
+
 
 def migrate_maas_dns():
     """

hooks/utils.py

@@ -1317,6 +1317,13 @@ def assess_status_helper():
             'blocked',
             'stonith_enabled config option is no longer supported')
 
+    if config('no_quorum_policy'):
+        if config('no_quorum_policy').lower() not in ['ignore', 'freeze',
+                                                      'stop', 'suicide']:
+            return(
+                'blocked',
+                'Invalid no_quorum_policy specified')
+
     if is_unit_upgrading_set():
         return ("blocked",
                 "Ready for do-release-upgrade. Set complete when finished")

unit_tests/test_hacluster_hooks.py

@@ -422,6 +422,7 @@ class TestHooks(test_utils.CharmTestCase):
         apt_install.assert_called_once_with(expected_pkgs, fatal=True)
         setup_ocf_files.assert_called_once_with()
 
+    @mock.patch('pcmk.set_property')
     @mock.patch.object(hooks, 'is_stonith_configured')
     @mock.patch.object(hooks, 'configure_stonith')
     @mock.patch.object(hooks, 'relation_ids')
@@ -429,6 +430,7 @@ class TestHooks(test_utils.CharmTestCase):
     @mock.patch.object(hooks, 'maintenance_mode')
     @mock.patch.object(hooks, 'is_leader')
     @mock.patch.object(hooks, 'update_nrpe_config')
+    @mock.patch.object(hooks, 'assess_status_helper')
     @mock.patch('pcmk.commit')
     @mock.patch('pcmk.wait_for_pcmk')
     @mock.patch.object(hooks, 'configure_corosync')
@@ -439,21 +441,25 @@ class TestHooks(test_utils.CharmTestCase):
     def test_config_changed(self, mock_mkdir, mock_rsync, mock_config,
                             mock_os_mkdir, mock_configure_corosync,
                             mock_wait_for_pcmk, mock_pcmk_commit,
+                            mock_assess_status_helper,
                             mock_update_nrpe_config, mock_is_leader,
                             mock_maintenance_mode,
                             mock_hanode_relation_joined,
                             mock_relation_ids,
                             mock_configure_stonith,
-                            mock_is_stonith_configured):
+                            mock_is_stonith_configured,
+                            mock_set_property):
 
         mock_is_stonith_configured.return_value = False
         mock_config.side_effect = self.test_config.get
         mock_relation_ids.return_value = ['hanode:1']
         mock_is_leader.return_value = True
+        mock_assess_status_helper.return_value = ('active', 'mockmessage')
         hooks.config_changed()
         mock_maintenance_mode.assert_not_called()
         mock_relation_ids.assert_called_with('hanode')
         mock_hanode_relation_joined.assert_called_once_with('hanode:1')
+        mock_set_property.assert_called_with('no-quorum-policy', 'stop')
 
         # enable maintenance
         self.test_config.set_previous('maintenance-mode', False)
@@ -467,6 +473,11 @@ class TestHooks(test_utils.CharmTestCase):
         hooks.config_changed()
         mock_maintenance_mode.assert_called_with(False)
 
+        # set no-quorum-policy to ignore
+        self.test_config.set('no_quorum_policy', 'ignore')
+        hooks.config_changed()
+        mock_set_property.assert_called_with('no-quorum-policy', 'ignore')
+
     @mock.patch.object(hooks, 'needs_maas_dns_migration')
     @mock.patch.object(hooks, 'relation_ids')
     def test_migrate_maas_dns_no_migration(self, relation_ids,