Skip to content

Add load_balance_hosts to spread connections across multiple hosts #1394

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
lesnik512 wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Add load_balance_hosts to spread connections across multiple hosts #1394

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 38 additions & 3 deletions asyncpg/connect_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,11 @@ class SSLNegotiation(compat.StrEnum):
direct = "direct"


class LoadBalanceHosts(compat.StrEnum):
disable = "disable"
random = "random"


_ConnectionParameters = collections.namedtuple(
'ConnectionParameters',
[
Expand All @@ -70,6 +75,7 @@ class SSLNegotiation(compat.StrEnum):
'ssl_negotiation',
'server_settings',
'target_session_attrs',
'load_balance_hosts',
'krbsrvname',
'gsslib',
])
Expand Down Expand Up @@ -283,7 +289,8 @@ def _parse_connect_dsn_and_args(*, dsn, host, port, user,
password, passfile, database, ssl,
service, servicefile,
direct_tls, server_settings,
target_session_attrs, krbsrvname, gsslib):
target_session_attrs, load_balance_hosts,
krbsrvname, gsslib):
# `auth_hosts` is the version of host information for the purposes
# of reading the pgpass file.
auth_hosts = None
Expand Down Expand Up @@ -431,6 +438,11 @@ def _parse_connect_dsn_and_args(*, dsn, host, port, user,
if target_session_attrs is None:
target_session_attrs = dsn_target_session_attrs

if 'load_balance_hosts' in query:
val = query.pop('load_balance_hosts')
if load_balance_hosts is None:
load_balance_hosts = val

if 'krbsrvname' in query:
val = query.pop('krbsrvname')
if krbsrvname is None:
Expand Down Expand Up @@ -548,6 +560,11 @@ def _parse_connect_dsn_and_args(*, dsn, host, port, user,
if target_session_attrs is None:
target_session_attrs = dsn_target_session_attrs

if 'load_balance_hosts' in service_params:
val = service_params.pop('load_balance_hosts')
if load_balance_hosts is None:
load_balance_hosts = val

if 'krbsrvname' in service_params:
val = service_params.pop('krbsrvname')
if krbsrvname is None:
Expand Down Expand Up @@ -854,6 +871,20 @@ def _parse_connect_dsn_and_args(*, dsn, host, port, user,
)
) from None

if load_balance_hosts is None:
load_balance_hosts = os.getenv(
'PGLOADBALANCEHOSTS', LoadBalanceHosts.disable
)
try:
load_balance_hosts = LoadBalanceHosts(load_balance_hosts)
except ValueError:
raise exceptions.ClientConfigurationError(
"load_balance_hosts is expected to be one of "
"{!r}, got {!r}".format(
[m.value for m in LoadBalanceHosts], load_balance_hosts
)
) from None

if krbsrvname is None:
krbsrvname = os.getenv('PGKRBSRVNAME')

Expand All @@ -871,6 +902,7 @@ def _parse_connect_dsn_and_args(*, dsn, host, port, user,
sslmode=sslmode, ssl_negotiation=sslneg,
server_settings=server_settings,
target_session_attrs=target_session_attrs,
load_balance_hosts=load_balance_hosts,
krbsrvname=krbsrvname, gsslib=gsslib)

return addrs, params
Expand All @@ -882,8 +914,8 @@ def _parse_connect_arguments(*, dsn, host, port, user, password, passfile,
max_cached_statement_lifetime,
max_cacheable_statement_size,
ssl, direct_tls, server_settings,
target_session_attrs, krbsrvname, gsslib,
service, servicefile):
target_session_attrs, load_balance_hosts,
krbsrvname, gsslib, service, servicefile):
local_vars = locals()
for var_name in {'max_cacheable_statement_size',
'max_cached_statement_lifetime',
Expand Down Expand Up @@ -913,6 +945,7 @@ def _parse_connect_arguments(*, dsn, host, port, user, password, passfile,
direct_tls=direct_tls, database=database,
server_settings=server_settings,
target_session_attrs=target_session_attrs,
load_balance_hosts=load_balance_hosts,
krbsrvname=krbsrvname, gsslib=gsslib,
service=service, servicefile=servicefile)

Expand Down Expand Up @@ -1268,6 +1301,8 @@ async def _connect(*, loop, connection_class, record_class, **kwargs):

addrs, params, config = _parse_connect_arguments(**kwargs)
target_attr = params.target_session_attrs
if params.load_balance_hosts is LoadBalanceHosts.random:
addrs = random.sample(addrs, len(addrs))

candidates = []
chosen_connection = None
Expand Down
17 changes: 17 additions & 0 deletions asyncpg/connection.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2131,6 +2131,7 @@ async def connect(dsn=None, *,
record_class=protocol.Record,
server_settings=None,
target_session_attrs=None,
load_balance_hosts=None,
krbsrvname=None,
gsslib=None):
r"""A coroutine to establish a connection to a PostgreSQL server.
Expand Down Expand Up @@ -2382,6 +2383,18 @@ async def connect(dsn=None, *,
or the value of the ``PGTARGETSESSIONATTRS`` environment variable,
or ``"any"`` if neither is specified.

:param str load_balance_hosts:
The order in which the addresses in *host* are tried. Can be one of:

- ``"disable"`` - in the order they are listed
- ``"random"`` - in a random order, chosen again on every
connection attempt, which spreads connections (for example,
those of a pool) across the hosts

If not specified, the value parsed from the *dsn* argument or the
service file is used, or the value of the ``PGLOADBALANCEHOSTS``
environment variable, or ``"disable"`` if neither is specified.

:param str krbsrvname:
Kerberos service name to use when authenticating with GSSAPI. This
must match the server configuration. Defaults to 'postgres'.
Expand Down Expand Up @@ -2472,6 +2485,9 @@ async def connect(dsn=None, *,
SSL contexts created by asyncpg now set the ``postgresql`` ALPN
protocol, which PostgreSQL 17+ requires for direct SSL connections.

.. versionchanged:: 0.33.0
Added the *load_balance_hosts* parameter.

.. _SSLContext: https://docs.python.org/3/library/ssl.html#ssl.SSLContext
.. _create_default_context:
https://docs.python.org/3/library/ssl.html#ssl.create_default_context
Expand Down Expand Up @@ -2516,6 +2532,7 @@ async def connect(dsn=None, *,
max_cached_statement_lifetime=max_cached_statement_lifetime,
max_cacheable_statement_size=max_cacheable_statement_size,
target_session_attrs=target_session_attrs,
load_balance_hosts=load_balance_hosts,
krbsrvname=krbsrvname,
gsslib=gsslib,
)
Expand Down
135 changes: 134 additions & 1 deletion tests/test_connect.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -751,6 +751,69 @@ class TestConnectParams(tb.TestCase):
})
},

{
'name': 'load_balance_hosts_default',
'dsn': 'postgresql://user@host1,host2/db',
'result': ([('host1', 5432), ('host2', 5432)], {
'database': 'db',
'user': 'user',
'target_session_attrs': 'any',
'load_balance_hosts': 'disable',
})
},

{
'name': 'load_balance_hosts_dsn',
'dsn': 'postgresql://user@host1,host2/db'
'?load_balance_hosts=random',
'env': {
'PGLOADBALANCEHOSTS': 'disable',
},
'result': ([('host1', 5432), ('host2', 5432)], {
'database': 'db',
'user': 'user',
'target_session_attrs': 'any',
'load_balance_hosts': 'random',
})
},

{
'name': 'load_balance_hosts_kwarg',
'dsn': 'postgresql://user@host1,host2/db'
'?load_balance_hosts=random',
'load_balance_hosts': 'disable',
'result': ([('host1', 5432), ('host2', 5432)], {
'database': 'db',
'user': 'user',
'target_session_attrs': 'any',
'load_balance_hosts': 'disable',
})
},

{
'name': 'load_balance_hosts_env',
'dsn': 'postgresql://user@host1,host2/db',
'env': {
'PGLOADBALANCEHOSTS': 'random',
},
'result': ([('host1', 5432), ('host2', 5432)], {
'database': 'db',
'user': 'user',
'target_session_attrs': 'any',
'load_balance_hosts': 'random',
})
},

{
'name': 'load_balance_hosts_invalid',
'dsn': 'postgresql://user@host1,host2/db'
'?load_balance_hosts=round-robin',
'error': (
exceptions.ClientConfigurationError,
'load_balance_hosts is expected to be one of'
)
},

{
'name': 'krbsrvname',
'dsn': 'postgresql://user@host/db?krbsrvname=srv_qs',
Expand Down Expand Up @@ -1144,7 +1207,7 @@ def run_testcase(self, testcase):
test_env = {'PGHOST': None, 'PGPORT': None,
'PGUSER': None, 'PGPASSWORD': None,
'PGDATABASE': None, 'PGSSLMODE': None,
'PGSERVICE': None, }
'PGSERVICE': None, 'PGLOADBALANCEHOSTS': None, }
test_env.update(env)

dsn = testcase.get('dsn')
Expand All @@ -1158,6 +1221,7 @@ def run_testcase(self, testcase):
direct_tls = testcase.get('direct_tls')
server_settings = testcase.get('server_settings')
target_session_attrs = testcase.get('target_session_attrs')
load_balance_hosts = testcase.get('load_balance_hosts')
krbsrvname = testcase.get('krbsrvname')
gsslib = testcase.get('gsslib')
service = testcase.get('service')
Expand Down Expand Up @@ -1187,6 +1251,7 @@ def run_testcase(self, testcase):
direct_tls=direct_tls,
server_settings=server_settings,
target_session_attrs=target_session_attrs,
load_balance_hosts=load_balance_hosts,
krbsrvname=krbsrvname, gsslib=gsslib,
service=service, servicefile=servicefile)

Expand Down Expand Up @@ -1218,6 +1283,8 @@ def run_testcase(self, testcase):
# Avoid the hassle of specifying gsslib
# unless explicitly tested for
params.pop('gsslib', None)
if 'load_balance_hosts' not in expected[1]:
params.pop('load_balance_hosts', None)

self.assertEqual(expected, result, 'Testcase: {}'.format(testcase))

Expand Down Expand Up @@ -1290,6 +1357,12 @@ def test_connect_connection_service_file(self):
target_session_attrs=primary
krbsrvname=fakekrbsrvname
gsslib=sspi

[test_service_load_balance_hosts]
host=somehost
dbname=test_dbname
user=admin
load_balance_hosts=random
'''))
connection_service_file.close()
os.chmod(connection_service_file.name, stat.S_IWUSR | stat.S_IRUSR)
Expand Down Expand Up @@ -1369,6 +1442,23 @@ def test_connect_connection_service_file(self):
}
)
})
# Test load_balance_hosts from the service file
self.run_testcase({
'dsn': 'postgresql://?service=test_service_load_balance_hosts',
'env': {
'PGSERVICEFILE': connection_service_file.name,
'PGLOADBALANCEHOSTS': 'disable',
},
'result': (
[('somehost', 5432)],
{
'user': 'admin',
'database': 'test_dbname',
'target_session_attrs': 'any',
'load_balance_hosts': 'random',
}
)
})
finally:
os.unlink(connection_service_file.name)

Expand Down Expand Up @@ -1758,6 +1848,49 @@ async def close():
await asyncio.gather(*pending, return_exceptions=True)


class TestLoadBalanceHosts(tb.TestCase):

async def connect_order(self, **kwargs):
tried = []

async def connect_addr(*, addr, **_):
tried.append(addr[0])
raise ConnectionRefusedError

with unittest.mock.patch.object(
connect_utils, '_connect_addr', side_effect=connect_addr,
):
with self.assertRaises(ConnectionRefusedError):
await asyncpg.connect(
host=['first', 'second', 'third'], user='postgres',
database='postgres', ssl=False, **kwargs)
return tried

async def test_load_balance_hosts_disable_keeps_order(self):
with unittest.mock.patch.object(
connect_utils.random, 'sample',
side_effect=lambda seq, k: list(reversed(seq)),
):
self.assertEqual(await self.connect_order(),
['first', 'second', 'third'])
self.assertEqual(
await self.connect_order(load_balance_hosts='disable'),
['first', 'second', 'third'])

async def test_load_balance_hosts_random_shuffles_each_connect(self):
orders = iter([[2, 0, 1], [1, 2, 0]])
with unittest.mock.patch.object(
connect_utils.random, 'sample',
side_effect=lambda seq, k: [seq[i] for i in next(orders)],
):
self.assertEqual(
await self.connect_order(load_balance_hosts='random'),
['third', 'first', 'second'])
self.assertEqual(
await self.connect_order(load_balance_hosts='random'),
['second', 'third', 'first'])


class TestConnection(tb.ConnectedTestCase):

async def test_connection_cancelled_during_transport_setup(self):
Expand Down