Add Etcd v3 protocol support via api gRPC-gateway (#1162)

The only python-etcd3 client working directly via gRPC still supports only a single endpoint, which is not very nice for high-availability.

Since Patroni is already using a heavily hacked version of python-etcd with smart retries and auto-discovery out-of-the-box, I decided to enhance the existing code with limited support of v3 protocol via gRPC-gateway.

Unfortunately, watches via gRPC-gateway requires us to open and keep the second connection to the etcd.

Known limitations:
* The very minimal supported version is 3.0.4. On earlier versions transactions don't work due to bugs in grpc-gateway. Without transactions we can't do atomic operations, i.e. leader locks.
* Watches work only starting from 3.1.0
* Authentication works only starting from 3.3.0
* gRPC-gateway does not support authentication using TLS Common Name. This is because gRPC-proxy terminates TLS from its client so all the clients share a cert of the proxy: https://github.com/etcd-io/etcd/blob/master/Documentation/op-guide/authentication.md#using-tls-common-name
This commit is contained in:
Alexander Kukushkin
2020-07-31 14:33:40 +02:00
committed by GitHub
parent b3c69b5fb2
commit 3341c898ff
20 changed files with 1271 additions and 125 deletions
+1
View File
@@ -169,6 +169,7 @@ class PostgresInit(unittest.TestCase):
'stats_temp_directory': '/tmp'}
@patch('psycopg2.connect', psycopg2_connect)
@patch('patroni.postgresql.CallbackExecutor', Mock())
@patch.object(ConfigHandler, 'write_postgresql_conf', Mock())
@patch.object(ConfigHandler, 'replace_pg_hba', Mock())
@patch.object(ConfigHandler, 'replace_pg_ident', Mock())
+2 -2
View File
@@ -8,7 +8,7 @@ from mock import patch, Mock
from patroni.ctl import ctl, store_config, load_config, output_members, get_dcs, parse_dcs, \
get_all_members, get_any_member, get_cursor, query_member, configure, PatroniCtlException, apply_config_changes, \
format_config_for_editing, show_diff, invoke_editor, format_pg_version, find_executable, CONFIG_FILE_PATH
from patroni.dcs.etcd import Client, Failover
from patroni.dcs.etcd import AbstractEtcdClientWithFailover, Failover
from patroni.utils import tzutc
from psycopg2 import OperationalError
from urllib3 import PoolManager
@@ -37,7 +37,7 @@ class TestCtl(unittest.TestCase):
@patch('socket.getaddrinfo', socket_getaddrinfo)
def setUp(self):
with patch.object(Client, 'machines') as mock_machines:
with patch.object(AbstractEtcdClientWithFailover, 'machines') as mock_machines:
mock_machines.__get__ = Mock(return_value=['http://remotehost:2379'])
self.runner = CliRunner()
self.e = get_dcs({'etcd': {'ttl': 30, 'host': 'ok:2379', 'retry_timeout': 10}}, 'foo')
+21 -18
View File
@@ -5,7 +5,7 @@ import unittest
from dns.exception import DNSException
from mock import Mock, patch
from patroni.dcs.etcd import AbstractDCS, Client, Cluster, Etcd, EtcdError, DnsCachingResolver
from patroni.dcs.etcd import AbstractDCS, EtcdClient, Cluster, Etcd, EtcdError, DnsCachingResolver
from patroni.exceptions import DCSError
from patroni.utils import Retry
from urllib3.exceptions import ReadTimeoutError
@@ -123,9 +123,9 @@ class TestClient(unittest.TestCase):
@patch('socket.getaddrinfo', socket_getaddrinfo)
@patch('patroni.dcs.etcd.requests_get', requests_get)
def setUp(self):
with patch.object(Client, 'machines') as mock_machines:
with patch.object(EtcdClient, 'machines') as mock_machines:
mock_machines.__get__ = Mock(return_value=['http://localhost:2379', 'http://localhost:4001'])
self.client = Client({'srv': 'test', 'retry_timeout': 3}, DnsCachingResolver())
self.client = EtcdClient({'srv': 'test', 'retry_timeout': 3}, DnsCachingResolver())
self.client.http.request = http_request
self.client.http.request_encode_body = http_request
@@ -143,10 +143,9 @@ class TestClient(unittest.TestCase):
except Exception:
self.assertIsNone(machines)
@patch.object(Client, 'machines')
@patch.object(EtcdClient, 'machines')
def test_api_execute(self, mock_machines):
mock_machines.__get__ = Mock(return_value=['http://localhost:4001', 'http://localhost:2379'])
self.assertRaises(ValueError, self.client.api_execute, '', '')
self.client._base_uri = 'http://localhost:4001'
self.assertRaises(etcd.EtcdException, self.client.api_execute, '/', 'POST', timeout=0)
self.client._base_uri = 'http://localhost:4001'
@@ -157,18 +156,17 @@ class TestClient(unittest.TestCase):
self.client._machines_cache = [self.client._base_uri]
self.assertRaises(etcd.EtcdWatchTimedOut, self.client.api_execute, '/timeout', 'POST', params={'wait': 'true'})
self.assertRaises(etcd.EtcdWatchTimedOut, self.client.api_execute, '/timeout', 'POST', params={'wait': 'true'})
self.assertRaises(etcd.EtcdException, self.client.api_execute, '/', '')
with patch.object(Client, '_calculate_timeouts', Mock(side_effect=[(1, 1, 0), (1, 1, 0), (0, 1, 0)])),\
patch.object(Client, '_load_machines_cache', Mock(side_effect=Exception)):
with patch.object(EtcdClient, '_calculate_timeouts', Mock(side_effect=[(1, 1, 0), (1, 1, 0), (0, 1, 0)])),\
patch.object(EtcdClient, '_load_machines_cache', Mock(side_effect=Exception)):
self.client.http.request = Mock(side_effect=socket.error)
self.assertRaises(etcd.EtcdException, rtry, self.client.api_execute, '/', 'GET', params={'retry': rtry})
with patch.object(Client, '_calculate_timeouts', Mock(side_effect=[(1, 1, 0), (1, 1, 0), (0, 1, 0)])),\
patch.object(Client, '_load_machines_cache', Mock(return_value=True)):
with patch.object(EtcdClient, '_calculate_timeouts', Mock(side_effect=[(1, 1, 0), (1, 1, 0), (0, 1, 0)])),\
patch.object(EtcdClient, '_load_machines_cache', Mock(return_value=True)):
self.assertRaises(etcd.EtcdException, rtry, self.client.api_execute, '/', 'GET', params={'retry': rtry})
with patch.object(Client, '_do_http_request', Mock(side_effect=etcd.EtcdException)):
with patch.object(EtcdClient, '_do_http_request', Mock(side_effect=etcd.EtcdException)):
self.client._read_timeout = 0.01
self.assertRaises(etcd.EtcdException, self.client.api_execute, '/', 'GET')
@@ -184,7 +182,7 @@ class TestClient(unittest.TestCase):
def test__get_machines_cache_from_dns(self):
self.client._get_machines_cache_from_dns('error', 2379)
@patch.object(Client, 'machines')
@patch.object(EtcdClient, 'machines')
def test__refresh_machines_cache(self, mock_machines):
mock_machines.__get__ = Mock(side_effect=etcd.EtcdConnectionFailed)
self.assertIsNone(self.client._refresh_machines_cache())
@@ -205,6 +203,10 @@ class TestClient(unittest.TestCase):
timeout=1, source_address=('localhost', 53333),
socket_options=[(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)])
def test___del__(self):
self.client.http.clear = Mock(side_effect=TypeError)
del self.client
@patch('patroni.dcs.etcd.requests_get', requests_get)
@patch('socket.getaddrinfo', socket_getaddrinfo)
@@ -215,7 +217,7 @@ class TestEtcd(unittest.TestCase):
@patch('socket.getaddrinfo', socket_getaddrinfo)
def setUp(self):
with patch.object(Client, 'machines') as mock_machines:
with patch.object(EtcdClient, 'machines') as mock_machines:
mock_machines.__get__ = Mock(return_value=['http://localhost:2379', 'http://localhost:4001'])
self.etcd = Etcd({'namespace': '/patroni/', 'ttl': 30, 'retry_timeout': 10,
'host': 'localhost:2379', 'scope': 'test', 'name': 'foo'})
@@ -226,17 +228,18 @@ class TestEtcd(unittest.TestCase):
@patch('dns.resolver.query', dns_query)
def test_get_etcd_client(self):
with patch('time.sleep', Mock(side_effect=SleepException)),\
patch.object(Client, 'machines') as mock_machines:
patch.object(EtcdClient, 'machines') as mock_machines:
mock_machines.__get__ = Mock(side_effect=etcd.EtcdException)
self.assertRaises(SleepException, self.etcd.get_etcd_client,
{'discovery_srv': 'test', 'retry_timeout': 10, 'cacert': '1', 'key': '1', 'cert': 1})
{'discovery_srv': 'test', 'retry_timeout': 10, 'cacert': '1', 'key': '1', 'cert': 1},
EtcdClient)
self.assertRaises(SleepException, self.etcd.get_etcd_client,
{'url': 'https://test:2379', 'retry_timeout': 10})
{'url': 'https://test:2379', 'retry_timeout': 10}, EtcdClient)
self.assertRaises(SleepException, self.etcd.get_etcd_client,
{'hosts': 'foo:4001,bar', 'retry_timeout': 10})
{'hosts': 'foo:4001,bar', 'retry_timeout': 10}, EtcdClient)
mock_machines.__get__ = Mock(return_value=[])
self.assertRaises(SleepException, self.etcd.get_etcd_client,
{'proxy': 'https://user:password@test:2379', 'retry_timeout': 10})
{'proxy': 'https://user:password@test:2379', 'retry_timeout': 10}, EtcdClient)
def test_get_cluster(self):
cluster = self.etcd.get_cluster()
+246
View File
@@ -0,0 +1,246 @@
import etcd
import json
import unittest
import urllib3
from mock import Mock, patch
from patroni.dcs.etcd3 import PatroniEtcd3Client, Cluster, Etcd3, Etcd3Error, Etcd3ClientError, RetryFailedError,\
InvalidAuthToken, Unavailable, Unknown, UnsupportedEtcdVersion, UserEmpty, base64_encode
from threading import Thread
from . import SleepException, MockResponse
def mock_urlopen(self, method, url, **kwargs):
ret = MockResponse()
if method == 'GET' and url.endswith('/version'):
ret.content = '{"etcdserver": "3.3.13", "etcdcluster": "3.3.0"}'
elif method != 'POST':
raise Exception('Unexpected request method: {0} {1} {2}'.format(method, url, kwargs))
elif url.endswith('/cluster/member/list'):
ret.content = '{"members":[{"clientURLs":["http://localhost:2379", "http://localhost:4001"]}]}'
elif url.endswith('/auth/authenticate'):
ret.content = '{"token":"authtoken"}'
elif url.endswith('/lease/grant'):
ret.content = '{"ID": "123"}'
elif url.endswith('/lease/keepalive'):
ret.content = '{"result":{"TTL":30}}'
elif url.endswith('/kv/range'):
ret.content = json.dumps({
"header": {"revision": "1"},
"kvs": [
{"key": base64_encode('/patroni/test/leader'),
"value": base64_encode('foo'), "lease": "bla", "mod_revision": '1'},
{"key": base64_encode('/patroni/test/members/foo'),
"value": base64_encode('{}'), "lease": "123", "mod_revision": '1'},
{"key": base64_encode('/patroni/test/failover'), "value": base64_encode('{}'), "mod_revision": '1'}
]
})
elif url.endswith('/watch'):
key = base64_encode('/patroni/test/config')
ret.read_chunked = Mock(return_value=[json.dumps({
'result': {'events': [
{'kv': {'key': key, 'value': base64_encode('bar'), 'mod_revision': '2'}},
{'kv': {'key': key, 'value': base64_encode('buzz'), 'mod_revision': '3'}},
{'type': 'DELETE', 'kv': {'key': key, 'mod_revision': '4'}},
{'kv': {'key': base64_encode('/patroni/test/optime/leader'),
'value': base64_encode('1234567'), 'mod_revision': '5'}},
]}
})[:-1].encode('utf-8'), b'}{"error":{"grpc_code":14,"message":"","http_code":503}}'])
elif url.endswith('/kv/put') or url.endswith('/kv/txn'):
ret.status_code = 400
ret.content = '{"code":5,"error":"etcdserver: requested lease not found"}'
elif not url.endswith('/kv/deleterange'):
raise Exception('Unexpected url: {0} {1} {2}'.format(method, url, kwargs))
return ret
class BaseTestEtcd3(unittest.TestCase):
@patch.object(Thread, 'start', Mock())
@patch.object(urllib3.PoolManager, 'urlopen', mock_urlopen)
def setUp(self):
self.etcd3 = Etcd3({'namespace': '/patroni/', 'ttl': 30, 'retry_timeout': 10,
'host': 'localhost:2378', 'scope': 'test', 'name': 'foo',
'username': 'etcduser', 'password': 'etcdpassword'})
self.client = self.etcd3._client
self.kv_cache = self.client._kv_cache
class TestKVCache(BaseTestEtcd3):
def test__do_watch(self):
self.client.watchprefix = Mock(return_value=False)
self.assertRaises(AttributeError, self.kv_cache._do_watch, '1')
@patch('time.sleep', Mock(side_effect=SleepException))
@patch('patroni.dcs.etcd3.KVCache._build_cache', Mock(side_effect=Exception))
def test_run(self):
self.assertRaises(SleepException, self.kv_cache.run)
@patch.object(urllib3.PoolManager, 'urlopen', mock_urlopen)
def test_kill_stream(self):
self.assertRaises(Unavailable, self.kv_cache._do_watch, '1')
self.kv_cache.kill_stream()
with patch.object(MockResponse, 'connection', create=True) as mock_conn:
self.kv_cache.kill_stream()
mock_conn.sock.close.side_effect = Exception
self.kv_cache.kill_stream()
class TestPatroniEtcd3Client(BaseTestEtcd3):
@patch('patroni.dcs.etcd3.Etcd3Client.authenticate', Mock(side_effect=Exception))
def test__init__(self):
self.assertRaises(SystemExit, self.setUp)
@patch.object(urllib3.PoolManager, 'urlopen')
def test_call_rpc(self, mock_urlopen):
request = {'key': base64_encode('/patroni/test/leader')}
mock_urlopen.return_value = MockResponse()
mock_urlopen.return_value.content = '{"succeeded":true,"header":{"revision":"1"}}'
self.client.call_rpc('/kv/txn', {'success': [{'request_delete_range': request}]})
self.client.call_rpc('/kv/put', request)
self.client.call_rpc('/kv/deleterange', request)
@patch('time.time', Mock(side_effect=[1, 10.9, 100]))
def test__wait_cache(self):
with self.kv_cache.condition:
self.assertRaises(RetryFailedError, self.client._wait_cache, 10)
@patch.object(urllib3.PoolManager, 'urlopen')
def test__restart_watcher(self, mock_urlopen):
mock_urlopen.return_value = MockResponse()
mock_urlopen.return_value.status_code = 400
mock_urlopen.return_value.content = '{"code":9,"error":"etcdserver: authentication is not enabled"}'
self.client.authenticate()
@patch.object(urllib3.PoolManager, 'urlopen')
def test__handle_auth_errors(self, mock_urlopen):
mock_urlopen.return_value = MockResponse()
mock_urlopen.return_value.content = '{"code":3,"error":"etcdserver: user name is empty"}'
mock_urlopen.return_value.status_code = 403
self.client._cluster_version = (3, 1, 5)
self.assertRaises(UnsupportedEtcdVersion, self.client.deleteprefix, 'foo')
self.client._cluster_version = (3, 3, 13)
self.assertRaises(UserEmpty, self.client.deleteprefix, 'foo')
mock_urlopen.return_value.content = '{"code":16,"error":"etcdserver: invalid auth token"}'
self.assertRaises(InvalidAuthToken, self.client.deleteprefix, 'foo')
with patch.object(PatroniEtcd3Client, 'authenticate', Mock(return_value=True)):
self.assertRaises(InvalidAuthToken, self.client.deleteprefix, 'foo')
self.client.username = None
self.assertRaises(InvalidAuthToken, self.client.deleteprefix, 'foo')
def test__handle_server_response(self):
response = MockResponse()
response.content = '{"code":0,"error":"'
self.assertRaises(etcd.EtcdException, self.client._handle_server_response, response)
response.status_code = 400
self.assertRaises(Unknown, self.client._handle_server_response, response)
response.content = '{"error":{"grpc_code":0,"message":"","http_code":400}}'
try:
self.client._handle_server_response(response)
except Unknown as e:
self.assertEqual(e.as_dict(), {'code': 2, 'codeText': 'OK', 'error': u'', 'status': 400})
@patch.object(urllib3.PoolManager, 'urlopen')
def test__ensure_version_prefix(self, mock_urlopen):
self.client.version_prefix = None
mock_urlopen.return_value = MockResponse()
mock_urlopen.return_value.content = '{"etcdserver": "3.0.3", "etcdcluster": "3.0.0"}'
self.assertRaises(UnsupportedEtcdVersion, self.client._ensure_version_prefix, '')
mock_urlopen.return_value.content = '{"etcdserver": "3.0.4", "etcdcluster": "3.0.0"}'
self.client._ensure_version_prefix('')
self.assertEqual(self.client.version_prefix, '/v3alpha')
mock_urlopen.return_value.content = '{"etcdserver": "3.4.4", "etcdcluster": "3.4.0"}'
self.client._ensure_version_prefix('')
self.assertEqual(self.client.version_prefix, '/v3')
@patch.object(urllib3.PoolManager, 'urlopen', mock_urlopen)
class TestEtcd3(BaseTestEtcd3):
@patch.object(Thread, 'start', Mock())
@patch.object(urllib3.PoolManager, 'urlopen', mock_urlopen)
def setUp(self):
super(TestEtcd3, self).setUp()
self.assertRaises(AttributeError, self.kv_cache._build_cache)
self.kv_cache._is_ready = True
self.etcd3.get_cluster()
def test_get_cluster(self):
self.assertIsInstance(self.etcd3.get_cluster(), Cluster)
self.client._kv_cache = None
with patch.object(urllib3.PoolManager, 'urlopen') as mock_urlopen:
mock_urlopen.side_effect = UnsupportedEtcdVersion('')
self.assertRaises(UnsupportedEtcdVersion, self.etcd3.get_cluster)
mock_urlopen.side_effect = SleepException()
self.assertRaises(Etcd3Error, self.etcd3.get_cluster)
def test_touch_member(self):
self.etcd3.touch_member({})
self.etcd3._lease = 'bla'
self.etcd3.touch_member({})
with patch.object(PatroniEtcd3Client, 'lease_grant', Mock(side_effect=Etcd3ClientError)):
self.etcd3.touch_member({})
def test__update_leader(self):
self.etcd3._lease = None
self.etcd3.update_leader('123')
self.etcd3.update_leader('124')
def test_attempt_to_acquire_leader(self):
self.etcd3._lease = None
self.assertFalse(self.etcd3.attempt_to_acquire_leader())
def test_set_ttl(self):
self.etcd3.set_ttl(20)
@patch.object(PatroniEtcd3Client, 'lease_keepalive', Mock(return_value=False))
def test_refresh_lease(self):
self.etcd3._last_lease_refresh = 0
self.etcd3.refresh_lease()
@patch('time.sleep', Mock(side_effect=SleepException))
@patch.object(PatroniEtcd3Client, 'lease_keepalive', Mock(return_value=False))
@patch.object(PatroniEtcd3Client, 'lease_grant', Mock(side_effect=Etcd3ClientError))
def test_create_lease(self):
self.etcd3._lease = None
self.etcd3._last_lease_refresh = 0
self.assertRaises(SleepException, self.etcd3.create_lease)
def test_set_failover_value(self):
self.etcd3.set_failover_value('', 1)
def test_set_config_value(self):
self.etcd3.set_config_value('')
def test_initialize(self):
self.etcd3.initialize()
def test_cancel_initialization(self):
self.etcd3.cancel_initialization()
def test_delete_leader(self):
self.etcd3.delete_leader()
def test_delete_cluster(self):
self.etcd3.delete_cluster()
def test_set_history_value(self):
self.etcd3.set_history_value('')
def test_set_sync_state_value(self):
self.etcd3.set_sync_state_value('')
def test_delete_sync_state(self):
self.etcd3.delete_sync_state()
def test_watch(self):
self.etcd3.set_ttl(10)
self.etcd3.watch(None, 0)
self.etcd3.watch(None, 0)
def test_set_socket_options(self):
with patch('socket.SIO_KEEPALIVE_VALS', 1, create=True):
self.etcd3.set_socket_options(Mock(), None)
+2 -2
View File
@@ -6,7 +6,7 @@ import sys
from mock import Mock, MagicMock, PropertyMock, patch, mock_open
from patroni.config import Config
from patroni.dcs import Cluster, ClusterConfig, Failover, Leader, Member, get_dcs, SyncState, TimelineHistory
from patroni.dcs.etcd import Client
from patroni.dcs.etcd import AbstractEtcdClientWithFailover
from patroni.exceptions import DCSError, PostgresConnectionException, PatroniFatalException
from patroni.ha import Ha, _MemberStatus
from patroni.postgresql import Postgresql
@@ -185,7 +185,7 @@ class TestHa(PostgresInit):
@patch.object(etcd.Client, 'read', etcd_read)
def setUp(self):
super(TestHa, self).setUp()
with patch.object(Client, 'machines') as mock_machines:
with patch.object(AbstractEtcdClientWithFailover, 'machines') as mock_machines:
mock_machines.__get__ = Mock(return_value=['http://remotehost:2379'])
self.p.set_state('running')
self.p.set_role('replica')
+3 -3
View File
@@ -9,7 +9,7 @@ import patroni.config as config
from mock import Mock, PropertyMock, patch
from patroni.api import RestApiServer
from patroni.async_executor import AsyncExecutor
from patroni.dcs.etcd import Client
from patroni.dcs.etcd import AbstractEtcdClientWithFailover
from patroni.exceptions import DCSError
from patroni.postgresql import Postgresql
from patroni.postgresql.config import ConfigHandler
@@ -53,7 +53,7 @@ class TestPatroni(unittest.TestCase):
@patch.object(BaseHTTPServer.HTTPServer, '__init__', Mock())
@patch.object(etcd.Client, 'read', etcd_read)
@patch.object(Thread, 'start', Mock())
@patch.object(Client, 'machines', PropertyMock(return_value=['http://remotehost:2379']))
@patch.object(AbstractEtcdClientWithFailover, 'machines', PropertyMock(return_value=['http://remotehost:2379']))
def setUp(self):
self._handlers = logging.getLogger().handlers[:]
RestApiServer._BaseServer__is_shut_down = Mock()
@@ -75,7 +75,7 @@ class TestPatroni(unittest.TestCase):
@patch('sys.argv', ['patroni.py', 'postgres0.yml'])
@patch('time.sleep', Mock(side_effect=SleepException))
@patch.object(etcd.Client, 'delete', Mock())
@patch.object(Client, 'machines', PropertyMock(return_value=['http://remotehost:2379']))
@patch.object(AbstractEtcdClientWithFailover, 'machines', PropertyMock(return_value=['http://remotehost:2379']))
@patch.object(Thread, 'join', Mock())
def test_patroni_patroni_main(self):
with patch('subprocess.call', Mock(return_value=1)):
-1
View File
@@ -98,7 +98,6 @@ class TestPostgresql(BaseTestPostgresql):
def setUp(self):
super(TestPostgresql, self).setUp()
self.p.config.write_postgresql_conf()
self.p._callback_executor = Mock()
@patch('subprocess.Popen')
@patch.object(Postgresql, 'wait_for_startup')