Merge branch 'master' of github.com:zalando/patroni into feature/disable-automatic-failover

patroni/ha.py

@@ -182,8 +182,8 @@ class Ha(object):
             is_master = json['role'] == 'master'
             xlog_location = None if is_master else json['xlog']['replayed_location']
             return (member, True, not is_master, xlog_location, json.get('tags', {}))
-        except:
+        except Exception as e:
-            logger.exception('request failed: GET %s', member.api_url)
+            logger.warning("request failed: GET %s (%s)", member.api_url, e)
         return (member, False, None, 0, {})
 
     def fetch_nodes_statuses(self, members):

patroni/postgresql.py

@@ -503,6 +503,12 @@ class Postgresql(object):
             self._state = value
 
     def start(self, block_callbacks=False):
+        # make sure we close all connections established against
+        # the former node, otherwise, we might get a stalled one
+        # after kill -9, which would report incorrect data to
+        # patroni.
+        self.close_connection()
+
         if self.is_running():
             logger.error('Cannot start PostgreSQL because one is already running.')
             return True
@@ -559,12 +565,6 @@ class Postgresql(object):
             return 'not accessible or not healty'
 
     def stop(self, mode='fast', block_callbacks=False, checkpoint=True):
-        # make sure we close all connections established against
-        # the former node, otherwise, we might get a stalled one
-        # after kill -9, which would report incorrect data to
-        # patroni.
-
-        self.close_connection()
         if not self.is_running():
             if not block_callbacks:
                 self.set_state('stopped')

postgres0.yml

@@ -4,10 +4,12 @@ name: postgresql0
 
 restapi:
   listen: 127.0.0.1:8008
+  connect_address: 127.0.0.1:8008
+#  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
+#  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
 #  authentication:
 #    username: username
 #    password: password
-  connect_address: 127.0.0.1:8008
 
 etcd:
   host: 127.0.0.1:4001
@@ -30,11 +32,11 @@ bootstrap:
 #        max_wal_senders: 5
 #        max_replication_slots: 5
 #        wal_log_hints: "on"
-        archive_mode: "on"
+#        archive_mode: "on"
-        archive_timeout: 1800s
+#        archive_timeout: 1800s
-        archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
+#        archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
-      recovery_conf:
+#      recovery_conf:
-        restore_command: cp ../wal_archive/%f %p
+#        restore_command: cp ../wal_archive/%f %p
 
   # some desired options for 'initdb'
   initdb:  # Note: It needs to be a list (some options need values, others are switches)

postgres1.yml

@@ -1,105 +1,77 @@
-ttl: &ttl 30
+scope: batman
-loop_wait: &loop_wait 10
+#namespace: /service/
-scope: &scope batman
+name: postgresql1
+
 restapi:
   listen: 127.0.0.1:8009
   connect_address: 127.0.0.1:8009
-#  auth: 'username:password'
 #  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
 #  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
+#  authentication:
+#    username: username
+#    password: password
+
 etcd:
-  scope: *scope
-  ttl: *ttl
   host: 127.0.0.1:4001
-  #discovery_srv: my-etcd.domain
+
-#consul:
+bootstrap:
-#  scope: *scope
+  # this section will be written into Etcd:/<namespace>/<scope>/config after initializing new cluster
-#  ttl: *ttl
+  # and all other cluster members will use it as a `global configuration`
-#  host: 127.0.0.1:8500
+  dcs:
-#zookeeper:
+    ttl: 30
-#  scope: *scope
+    loop_wait: 10
-#  session_timeout: *ttl
+    retry_timeout: 10
-#  reconnect_timeout: *loop_wait
+    maximum_lag_on_failover: 1048576
-#  hosts:
+    postgresql:
-#    - 127.0.0.1:2181
+      use_pg_rewind: true
-#    - 127.0.0.2:2181
+#      use_slots: true
-#exhibitor:
+      parameters:
-#  poll_interval: 300
+#        wal_level: hot_standby
-#  port: 8181
+#        hot_standby: "on"
-#  hosts:
+#        wal_keep_segments: 8
-#    - host1
+#        max_wal_senders: 5
-#    - host2
+#        max_replication_slots: 5
-#    - host3
+#        wal_log_hints: "on"
+#        archive_mode: "on"
+#        archive_timeout: 1800s
+#        archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
+#      recovery_conf:
+#        restore_command: cp ../wal_archive/%f %p
+
+  # some desired options for 'initdb'
+  initdb:  # Note: It needs to be a list (some options need values, others are switches)
+  - encoding: UTF8
+  - data-checksums
+
+  pg_hba:  # Add following lines to pg_hba.conf after running 'initdb'
+  - host replication replicator 127.0.0.1/32 md5
+  - host all all 0.0.0.0/0 md5
+#  - hostssl all all 0.0.0.0/0 md5
+
+  # Some additional users users which needs to be created after initializing new cluster
+  users:
+    admin:
+      password: admin
+      options:
+        - createrole
+        - createdb
+
 postgresql:
-  name: postgresql1
-  scope: *scope
   listen: 127.0.0.1:5433
   connect_address: 127.0.0.1:5433
   data_dir: data/postgresql1
-  maximum_lag_on_failover: 1048576 # 1 megabyte in bytes
+#  bin_dir:
-  use_slots: True
   pgpass: /tmp/pgpass1
-  initdb:  ## We allow the following options to be passed on to initdb
+  authentication:
-  # - auth: authmethod
+    replication:
-  # - auth-host: authmethod
+      username: replicator
-  # - auth-local: authmethod
+      password: rep-pass
-  - encoding: UTF8
+    superuser:
-  # - data-checksums  # When pg_rewind is needed on 9.3, this needs to be enabled
+      username: postgres
-  # - locale: locale
+      password: zalando
-  # - lc-collate: locale
-  # - lc-ctype: locale
-  # - lc-messages: locale
-  # - lc-monetary: locale
-  # - lc-numeric: locale
-  # - lc-time: locale
-  # - text-search-config: CFG
-  # - xlogdir: directory
-  # - debug
-  # - noclean
-  pg_rewind:
-    username: postgres
-    password: zalando
-  pg_hba:
-  - host replication replicator 127.0.0.1/32 md5
-  - host all all 0.0.0.0/0 md5
-  # - hostssl all all 0.0.0.0/0 md5
-  replication:
-    username: replicator
-    password: rep-pass
-  superuser:
-    username: postgres
-    password: zalando
-  admin:
-    username: admin
-    password: admin
-# commented-out example for wal-e provisioning
-  create_replica_method:
-    - basebackup
-#   - wal_e
-# commented-out example for wal-e provisioning
-  #wal_e:
-    #command: /patroni/scripts/wale_restore.py
-    #env_dir: /home/postgres/etc/wal-e.d/env
-    #threshold_megabytes: 10240
-    #threshold_backup_size_percentage: 30
-    #retries: 2
-    #use_iam: 1
-  #recovery_conf:
-    #restore_command: envdir /etc/wal-e.d/env wal-e wal-fetch "%f" "%p" -p 1
-  recovery_conf:
-    restore_command: cp ../wal_archive/%f %p
   parameters:
-    archive_mode: "on"
-    wal_level: hot_standby
-    archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
-    max_wal_senders: 10
-    wal_keep_segments: 8
-    archive_timeout: 1800s
-    max_replication_slots: 10
-    hot_standby: "on"
-    wal_log_hints: "on"
     unix_socket_directories: '.'
 tags:
-    nofailover: False
+    nofailover: false
-    noloadbalance: False
+    noloadbalance: false
-    clonefrom: False
+    clonefrom: false

postgres2.yml

@@ -1,106 +1,78 @@
-ttl: &ttl 30
+scope: batman
-loop_wait: &loop_wait 10
+#namespace: /service/
-scope: &scope batman
+name: postgresql2
+
 restapi:
   listen: 127.0.0.1:8010
   connect_address: 127.0.0.1:8010
-  auth: 'username:password'
 #  certfile: /etc/ssl/certs/ssl-cert-snakeoil.pem
 #  keyfile: /etc/ssl/private/ssl-cert-snakeoil.key
+  authentication:
+    username: username
+    password: password
+
 etcd:
-  scope: *scope
-  ttl: *ttl
   host: 127.0.0.1:4001
-  #discovery_srv: my-etcd.domain
+
-#consul:
+bootstrap:
-#  scope: *scope
+  # this section will be written into Etcd:/<namespace>/<scope>/config after initializing new cluster
-#  ttl: *ttl
+  # and all other cluster members will use it as a `global configuration`
-#  host: 127.0.0.1:8500
+  dcs:
-#zookeeper:
+    ttl: 30
-#  scope: *scope
+    loop_wait: 10
-#  session_timeout: *ttl
+    retry_timeout: 10
-#  reconnect_timeout: *loop_wait
+    maximum_lag_on_failover: 1048576
-#  hosts:
+    postgresql:
-#    - 127.0.0.1:2181
+      use_pg_rewind: true
-#    - 127.0.0.2:2181
+#      use_slots: true
-#exhibitor:
+      parameters:
-#  poll_interval: 300
+#        wal_level: hot_standby
-#  port: 8181
+#        hot_standby: "on"
-#  hosts:
+#        wal_keep_segments: 8
-#    - host1
+#        max_wal_senders: 5
-#    - host2
+#        max_replication_slots: 5
-#    - host3
+#        wal_log_hints: "on"
+#        archive_mode: "on"
+#        archive_timeout: 1800s
+#        archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
+#      recovery_conf:
+#        restore_command: cp ../wal_archive/%f %p
+
+  # some desired options for 'initdb'
+  initdb:  # Note: It needs to be a list (some options need values, others are switches)
+  - encoding: UTF8
+  - data-checksums
+
+  pg_hba:  # Add following lines to pg_hba.conf after running 'initdb'
+  - host replication replicator 127.0.0.1/32 md5
+  - host all all 0.0.0.0/0 md5
+#  - hostssl all all 0.0.0.0/0 md5
+
+  # Some additional users users which needs to be created after initializing new cluster
+  users:
+    admin:
+      password: admin
+      options:
+        - createrole
+        - createdb
+
 postgresql:
-  name: postgresql2
-  scope: *scope
   listen: 127.0.0.1:5434
   connect_address: 127.0.0.1:5434
   data_dir: data/postgresql2
-  maximum_lag_on_failover: 1048576 # 1 megabyte in bytes
+#  bin_dir:
-  use_slots: True
   pgpass: /tmp/pgpass2
-  initdb:  ## We allow the following options to be passed on to initdb
+  authentication:
-  # - auth: authmethod
+    replication:
-  # - auth-host: authmethod
+      username: replicator
-  # - auth-local: authmethod
+      password: rep-pass
-  - encoding: UTF8
+    superuser:
-  # - data-checksums  # When pg_rewind is needed on 9.3, this needs to be enabled
+      username: postgres
-  # - locale: locale
+      password: zalando
-  # - lc-collate: locale
-  # - lc-ctype: locale
-  # - lc-messages: locale
-  # - lc-monetary: locale
-  # - lc-numeric: locale
-  # - lc-time: locale
-  # - text-search-config: CFG
-  # - xlogdir: directory
-  # - debug
-  # - noclean
-  pg_rewind:
-    username: postgres
-    password: zalando
-  pg_hba:
-  - host replication replicator 127.0.0.1/32 md5
-  - host all all 0.0.0.0/0 md5
-  # - hostssl all all 0.0.0.0/0 md5
-  replication:
-    username: replicator
-    password: rep-pass
-  superuser:
-    username: postgres
-    password: zalando
-  admin:
-    username: admin
-    password: admin
-# commented-out example for wal-e provisioning
-  create_replica_method:
-    - basebackup
-#   - wal_e
-# commented-out example for wal-e provisioning
-  #wal_e:
-    #command: /patroni/scripts/wale_restore.py
-    #env_dir: /home/postgres/etc/wal-e.d/env
-    #threshold_megabytes: 10240
-    #threshold_backup_size_percentage: 30
-    #retries: 2
-    #use_iam: 1
-  #recovery_conf:
-    #restore_command: envdir /etc/wal-e.d/env wal-e wal-fetch "%f" "%p" -p 1
-  recovery_conf:
-    restore_command: cp ../wal_archive/%f %p
   parameters:
-    archive_mode: "on"
-    wal_level: hot_standby
-    archive_command: mkdir -p ../wal_archive && test ! -f ../wal_archive/%f && cp %p ../wal_archive/%f
-    max_wal_senders: 10
-    wal_keep_segments: 8
-    archive_timeout: 1800s
-    max_replication_slots: 10
-    hot_standby: "on"
-    wal_log_hints: "on"
     unix_socket_directories: '.'
 tags:
-    nofailover: False
+    nofailover: false
-    noloadbalance: False
+    noloadbalance: false
-    clonefrom: False
+    clonefrom: false
-    replicatefrom: postgresql1
+    replicatefrom: postgres1

tests/test_api.py

@@ -188,7 +188,8 @@ class TestRestApiHandler(unittest.TestCase):
 
         post = request + '\nContent-Length: '
 
-        def make_request(request):
+        def make_request(request=None, **kwargs):
+            request = json.dumps(kwargs) if request is None else request
             return '{0}{1}\n\n{2}'.format(post, len(request), request)
 
         # empty request
@@ -198,26 +199,26 @@ class TestRestApiHandler(unittest.TestCase):
         request = make_request('foobar=baz')
         MockRestApiServer(RestApiHandler, request)
         # wrong role
-        request = make_request('{"schedule": "2016-08-20 12:45TZ+1", "role": "unknown", "postgres_version": "9.5.3"}')
+        request = make_request(schedule=future_restart_time.isoformat(), role='unknown', postgres_version='9.5.3')
         MockRestApiServer(RestApiHandler, request)
         # wrong version
-        request = make_request('{"schedule": "2016-08-20 12:45TZ+1", "role": "master", "postgres_version": "9.5.3.1"}')
+        request = make_request(schedule=future_restart_time.isoformat(), role='master', postgres_version='9.5.3.1')
         MockRestApiServer(RestApiHandler, request)
         # unknown filter
-        request = make_request('{"schedule": "2016-08-29 12:45TZ+1", "batman": "lives"}')
+        request = make_request(schedule=future_restart_time.isoformat(), batman='lives')
         MockRestApiServer(RestApiHandler, request)
         # incorrect schedule
-        request = make_request('{"schedule": "2016-08-42 12:45TZ+1", "role": "master"}')
+        request = make_request(schedule='2016-08-42 12:45TZ+1', role='master')
         MockRestApiServer(RestApiHandler, request)
         # everything fine, but the schedule is missing
-        request = make_request('{"role": "master", "postgres_version": "9.5.2"}')
+        request = make_request(role='master', postgres_version='9.5.2')
         MockRestApiServer(RestApiHandler, request)
         for retval in (True, False):
             with patch.object(MockHa, 'schedule_future_restart', Mock(return_value=retval)):
-                request = make_request('{"schedule": "2016-08-29 12:45TZ+1"}')
+                request = make_request(schedule=future_restart_time.isoformat())
                 MockRestApiServer(RestApiHandler, request)
             with patch.object(MockHa, 'restart', Mock(return_value=(retval, "foo"))):
-                request = make_request('{"role": "master", "postgres_version": "9.5.2"}')
+                request = make_request(role='master', postgres_version='9.5.2')
                 MockRestApiServer(RestApiHandler, request)
 
     def test_do_DELETE_restart(self):