fix(client/python): reachable cert-validation flag; bounded off-loop TOFU probe; license/marker fixes (Client.Python-027..031)
This commit is contained in:
@@ -12,6 +12,10 @@ import grpc
|
||||
from .auth import REDACTED, ApiKey
|
||||
from .errors import MxGatewayTransportError
|
||||
|
||||
# Fallback bound for the TOFU certificate probe when no call_timeout is set, so a
|
||||
# black-holed host fails fast instead of hanging on the OS default connect timeout.
|
||||
_TOFU_PROBE_TIMEOUT_SECONDS = 10.0
|
||||
|
||||
|
||||
@dataclass(frozen=True)
|
||||
class ClientOptions:
|
||||
@@ -88,8 +92,17 @@ def _split_authority(endpoint: str) -> tuple[str, int]:
|
||||
remainder = target[bracket_end + 1 :] # ":5120" or ""
|
||||
port_str = remainder.lstrip(":")
|
||||
return (host, int(port_str) if port_str else 443)
|
||||
host, _, port = target.rpartition(":")
|
||||
return (host or "localhost", int(port) if port else 443)
|
||||
host, sep, port = target.rpartition(":")
|
||||
if not sep:
|
||||
# No colon at all (e.g. a bare hostname "mygateway"): the whole target
|
||||
# is the host; default the port rather than raising on int("mygateway").
|
||||
return (target or "localhost", 443)
|
||||
if not port.isdigit():
|
||||
# A colon with a non-numeric / empty tail (e.g. a trailing ":") is not
|
||||
# an explicit port — keep the left side as the host and default the
|
||||
# port so a typo cannot raise an uncaught ValueError on the TOFU path.
|
||||
return (host or "localhost", 443)
|
||||
return (host or "localhost", int(port))
|
||||
|
||||
|
||||
def create_channel(options: ClientOptions) -> grpc.aio.Channel:
|
||||
@@ -120,9 +133,15 @@ def create_channel(options: ClientOptions) -> grpc.aio.Channel:
|
||||
else:
|
||||
# Lenient default: grpc-python has no per-channel skip-verify, so fetch the
|
||||
# server's certificate (unverified) and pin it for this channel (TOFU).
|
||||
# The probe opens a real blocking TCP+TLS socket, so it MUST be bounded —
|
||||
# a black-holed / firewall-drop host would otherwise hang on the OS default
|
||||
# connect timeout (minutes). Bound it by call_timeout (or a short fixed
|
||||
# fallback) so the dial fails fast as a transport error. The async
|
||||
# `connect` classmethods run this off the event loop (asyncio.to_thread).
|
||||
host, port = _split_authority(options.endpoint)
|
||||
probe_timeout = options.call_timeout if options.call_timeout else _TOFU_PROBE_TIMEOUT_SECONDS
|
||||
try:
|
||||
presented = ssl.get_server_certificate((host, port))
|
||||
presented = ssl.get_server_certificate((host, port), timeout=probe_timeout)
|
||||
except OSError as error:
|
||||
raise MxGatewayTransportError(
|
||||
f"failed to fetch TLS certificate from {options.endpoint}: {error}"
|
||||
|
||||
Reference in New Issue
Block a user