hermes-agent/agent/relay_runtime.py

1465 lines
55 KiB
Python

"""Profile-scoped NeMo Relay runtimes owned by the Hermes agent core."""
from __future__ import annotations
import atexit
import asyncio
import contextvars
import importlib
import inspect
import logging
import threading
import uuid
from concurrent.futures import TimeoutError as FuturesTimeoutError
from dataclasses import dataclass, field
from typing import Any, Callable
from hermes_constants import get_hermes_home
logger = logging.getLogger(__name__)
SESSION_SCOPE = "hermes.session"
TURN_SCOPE = "hermes.turn"
LOGICAL_LLM_SCOPE = "hermes.logical_llm_call"
RUNTIME_SCHEMA_KEY = "hermes.relay.schema_version"
RUNTIME_SCHEMA_VERSION = "hermes.relay.runtime.v1"
RUNTIME_INSTANCE_KEY = "hermes.relay.runtime_instance"
_PROFILE_KEY_CACHE: dict[str, str] = {}
# Bound for native scope lifecycle operations (push/pop/flush) that gate
# turn/session completion. Healthy operations complete in microseconds;
# only a wedged native pipeline breaches this, and the correct trade there
# is one lost span, never a blocked agent (2026-08-10 delegation stall).
_SCOPE_OP_TIMEOUT = 10.0
_SCOPE_OP_EXECUTOR: Any = None
_SCOPE_OP_EXECUTOR_LOCK = threading.Lock()
def _scope_op_executor():
"""Shared daemon executor for bounded native scope operations.
Daemon workers (tools.daemon_pool) so a wedged native call abandoned at
timeout cannot block interpreter exit. Sized generously: workers are
only consumed for the duration of healthy (microsecond) operations plus
any wedged calls, and ``Future.result(timeout=...)`` bounds callers even
when every worker is consumed by wedged calls — an unstarted future
still honors the result timeout, so exhaustion degrades to fast
timeouts, never a new hang.
"""
global _SCOPE_OP_EXECUTOR
if _SCOPE_OP_EXECUTOR is None:
with _SCOPE_OP_EXECUTOR_LOCK:
if _SCOPE_OP_EXECUTOR is None:
from tools.daemon_pool import DaemonThreadPoolExecutor
_SCOPE_OP_EXECUTOR = DaemonThreadPoolExecutor(
max_workers=8,
thread_name_prefix="relay-scope-op",
)
return _SCOPE_OP_EXECUTOR
def _run_bounded_on_exit_thread(fn: Callable[[], Any], timeout: float) -> Any:
"""Bounded fallback lane for interpreter shutdown.
When the shared executor refuses new futures (interpreter shutdown),
the operation still must not run unbounded on the calling thread: a
wedged native call would block process exit forever — the same defect
class this module exists to prevent, on the exit lane. Run it on a
fresh daemon thread with a bounded join; on breach the daemon worker
is abandoned exactly like the executor lane abandons its worker.
"""
result: list[Any] = []
error: list[BaseException] = []
def _target() -> None:
try:
result.append(fn())
except BaseException as exc: # noqa: BLE001 - propagated below
error.append(exc)
worker = threading.Thread(
target=_target, daemon=True, name="relay-scope-op-exit"
)
worker.start()
worker.join(timeout)
if worker.is_alive():
raise TimeoutError(
f"Relay scope operation exceeded {timeout}s during interpreter "
"shutdown; abandoning the native call so process exit can proceed"
)
if error:
raise error[0]
return result[0] if result else None
@dataclass
class RelaySession:
"""One isolated Relay scope stack owned by a Hermes session."""
session_id: str
parent_session_id: str = ""
lock: threading.RLock = field(default_factory=threading.RLock, repr=False)
closing: bool = False
handle: Any = None
context: contextvars.Context | None = None
# --- session-span segmentation (continuous sessions) ---
# Segment index of the CURRENT session scope (0 = first). Rotation
# closes the current scope and pushes segment N+1 at a turn boundary.
segment: int = 0
# Turns completed within the current segment (max_turns accounting).
segment_turns: int = 0
# Set by compaction notification; consumed at the next begin_turn.
rotate_pending: bool = False
# Rotating compaction landed while a turn was live on THIS session:
# closing now would pop the session scope under a live turn scope
# (LIFO violation). end_turn consumes this and closes the session.
close_pending: bool = False
# ---------------------------------------------------------------------------
# Session-span segmentation config (gateway.telemetry.session_segments).
# Cached at first read; both defaults OFF => rotation never fires and the
# scope lifecycle is byte-identical to the pre-segmentation behavior.
# ---------------------------------------------------------------------------
_SEGMENTS_CONFIG: dict[str, Any] | None = None
_SEGMENTS_CONFIG_LOCK = threading.Lock()
def _segments_config() -> dict[str, Any]:
"""Resolve session-segmentation settings; inert defaults when unset."""
global _SEGMENTS_CONFIG
if _SEGMENTS_CONFIG is None:
with _SEGMENTS_CONFIG_LOCK:
if _SEGMENTS_CONFIG is None:
on_compaction = False
max_turns = 0
try:
from gateway.run import _load_gateway_config # late import
telemetry = (
(_load_gateway_config().get("gateway") or {}).get(
"telemetry"
)
or {}
)
segments = telemetry.get("session_segments") or {}
on_compaction = bool(segments.get("on_compaction", False))
try:
max_turns = max(0, int(segments.get("max_turns", 0) or 0))
except (TypeError, ValueError):
max_turns = 0
except Exception: # noqa: BLE001 - config absence must not crash
pass
_SEGMENTS_CONFIG = {
"on_compaction": on_compaction,
"max_turns": max_turns,
}
return _SEGMENTS_CONFIG
def _reset_segments_config_for_tests() -> None:
global _SEGMENTS_CONFIG
_SEGMENTS_CONFIG = None
class RelayRuntime:
"""Own Relay session scopes independently of any exporter or plugin."""
def __init__(self, relay: Any = None, *, profile_key: str | None = None) -> None:
self.relay = relay or _load_nemo_relay()
self.profile_key = profile_key or current_profile_key()
self.runtime_id = uuid.uuid4().hex
self._sessions_lock = threading.RLock()
self._sessions: dict[str, RelaySession] = {}
self._subagent_parents: dict[str, str] = {}
self._subagent_parent_handles: dict[str, Any] = {}
self._execution_consumers_lock = threading.RLock()
self._execution_consumers: set[str] = set()
self._shutdown_registered = True
atexit.register(self.shutdown)
def retain_managed_execution(self, consumer: str) -> None:
"""Keep managed LLM and tool execution active for one consumer."""
if not consumer:
raise ValueError("Relay managed-execution consumer must not be empty")
with self._execution_consumers_lock:
self._execution_consumers.add(consumer)
def release_managed_execution(self, consumer: str) -> None:
"""Release a consumer's managed-execution requirement."""
with self._execution_consumers_lock:
self._execution_consumers.discard(consumer)
def managed_execution_enabled(self) -> bool:
"""Return whether a Hermes-managed consumer needs the Relay pipeline."""
with self._execution_consumers_lock:
return bool(self._execution_consumers)
def ensure_session(
self,
event: dict[str, Any],
*,
data: Any = None,
metadata: dict[str, Any] | None = None,
) -> RelaySession | None:
"""Return the existing session scope or create it once."""
session_id = _session_id(event)
if not session_id:
return None
with self._sessions_lock:
session = self._sessions.get(session_id)
if session is None:
parent_session_id = self._subagent_parents.get(session_id, "")
session = RelaySession(
session_id=session_id,
parent_session_id=parent_session_id,
)
self._sessions[session_id] = session
with session.lock:
if session.closing:
return None
if session.handle is None:
parent_handle = None
scope_metadata = {
**(metadata or {}),
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: self.runtime_id,
}
if session.parent_session_id:
with self._sessions_lock:
parent_handle = self._subagent_parent_handles.get(session_id)
if parent_handle is None:
parent = self.ensure_session({
"session_id": session.parent_session_id
})
if parent is not None:
parent_handle = parent.handle
scope_metadata["nemo_relay_scope_role"] = "subagent"
context = contextvars.Context()
try:
try:
session.handle = _scope_op_executor().submit(
context.run,
self.relay.scope.push,
SESSION_SCOPE,
self.relay.ScopeType.Agent,
handle=parent_handle,
data=data,
input={},
metadata=scope_metadata,
).result(timeout=_SCOPE_OP_TIMEOUT)
except RuntimeError:
# Interpreter shutdown: executor refuses new futures;
# push synchronously (no agent turn waits at exit).
session.handle = context.run(
self.relay.scope.push,
SESSION_SCOPE,
self.relay.ScopeType.Agent,
handle=parent_handle,
data=data,
input={},
metadata=scope_metadata,
)
except Exception:
session.context = None
raise
session.context = context
return session
def rotate_session_scope(self, session: RelaySession, *, reason: str) -> None:
"""Close the current session scope and open the next segment.
Called ONLY at a turn boundary (before the turn scope pushes), never
mid-turn: the scope stack is LIFO and rotating under a live child
would close a parent out of order. Both native calls are bounded by
``_SCOPE_OP_TIMEOUT`` — a wedged rotation costs one segment span,
never the agent. Segment bookkeeping advances even when a native
call fails, so a degraded rotation cannot retry on every turn.
"""
with session.lock:
if session.closing or session.handle is None:
return
old_handle = session.handle
# Advance bookkeeping FIRST: a failed native call must not leave
# rotate_pending set (tight rotation loop on every turn).
session.segment += 1
session.segment_turns = 0
session.rotate_pending = False
try:
self.run_in_session(
session,
self.relay.scope.pop,
old_handle,
output={"hermes.session.segment_reason": reason},
metadata={
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: self.runtime_id,
},
timeout=_SCOPE_OP_TIMEOUT,
)
except Exception:
logger.warning(
"Hermes Relay segment close failed (session=%s segment=%d); "
"abandoning the old segment span",
session.session_id,
session.segment - 1,
exc_info=True,
)
scope_metadata = {
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: self.runtime_id,
"hermes.session.segment": session.segment,
"hermes.session.segment_reason": reason,
}
parent_handle = None
if session.parent_session_id:
with self._sessions_lock:
parent_handle = self._subagent_parent_handles.get(
session.session_id
)
scope_metadata["nemo_relay_scope_role"] = "subagent"
context = contextvars.Context()
try:
session.handle = _scope_op_executor().submit(
context.run,
self.relay.scope.push,
SESSION_SCOPE,
self.relay.ScopeType.Agent,
handle=parent_handle,
input={},
metadata=scope_metadata,
).result(timeout=_SCOPE_OP_TIMEOUT)
session.context = context
except Exception:
logger.warning(
"Hermes Relay segment open failed (session=%s segment=%d); "
"keeping the prior scope handle",
session.session_id,
session.segment,
exc_info=True,
)
def register_subagent(
self,
event: dict[str, Any],
*,
metadata: dict[str, Any] | None = None,
) -> RelaySession | None:
"""Open a child Agent scope under its spawning turn when available."""
parent_session_id = str(event.get("parent_session_id") or "")
child_session_id = str(event.get("child_session_id") or "")
if (
not parent_session_id
or not child_session_id
or parent_session_id == child_session_id
):
return None
parent = self.ensure_session({"session_id": parent_session_id})
parent_handle = None if parent is None else parent.handle
turn = active_turn(parent_session_id)
if (
turn is not None
and not turn.closed
and turn.handle is not None
and turn.lease.host is self
and turn.lease.session is not None
and turn.lease.session.session_id == parent_session_id
):
parent_handle = turn.handle
with self._sessions_lock:
self._subagent_parents[child_session_id] = parent_session_id
if parent_handle is not None:
self._subagent_parent_handles[child_session_id] = parent_handle
return self.ensure_session(
{"session_id": child_session_id},
metadata=metadata,
)
def unregister_subagent(self, event: dict[str, Any]) -> None:
"""Close a delegated session and forget its parent relationship."""
child_session_id = str(event.get("child_session_id") or "")
if not child_session_id:
return
self.close_session({"session_id": child_session_id})
with self._sessions_lock:
self._subagent_parents.pop(child_session_id, None)
self._subagent_parent_handles.pop(child_session_id, None)
def get_session(self, session_id: str) -> RelaySession | None:
"""Return an active Hermes Relay session without creating one."""
with self._sessions_lock:
session = self._sessions.get(str(session_id or ""))
if session is None:
return None
with session.lock:
return None if session.closing else session
def get_session_handle(self, session_id: str) -> Any:
"""Return the Relay parent handle for a Hermes session, if active."""
session = self.get_session(session_id)
return None if session is None else session.handle
def run_in_session(
self,
session: RelaySession,
callback: Callable[..., Any],
*args: Any,
allow_closing: bool = False,
timeout: float | None = None,
**kwargs: Any,
) -> Any:
"""Run a Relay operation against a session's isolated scope stack.
``timeout`` (seconds) bounds the native call by running it on a
shared daemon executor; ``TimeoutError`` propagates to the caller's
existing exception handling on breach. ``None`` (default) preserves
the historical synchronous behavior. Scope lifecycle operations
that gate turn/session completion pass ``_SCOPE_OP_TIMEOUT``: the
native binding's ``scope.pop`` "returns after the scope is closed
successfully" — unbounded — and a wedged native pipeline (proven
live 2026-08-10 in the delegation topology) must cost at most one
span, never the agent. The abandoned daemon worker cannot block
process exit (tools.daemon_pool contract).
"""
with session.lock:
if session.closing and not allow_closing:
raise RuntimeError("Hermes Relay session is closing")
if session.context is None or session.handle is None:
raise RuntimeError("Hermes Relay session context is unavailable")
relay_context = session.context.copy()
context = contextvars.copy_context()
for variable, value in relay_context.items():
context.run(variable.set, value)
def invoke() -> Any:
self.relay.get_scope_stack()
return callback(*args, **kwargs)
# A copy permits a helper called by an existing Relay callback to
# re-enter the same logical session without re-entering Context.
if timeout is None:
return context.run(invoke)
try:
future = _scope_op_executor().submit(context.run, invoke)
except RuntimeError:
# Interpreter shutdown: the executor refuses new futures, but
# the atexit close path must still flush cleanly. Still
# bounded — a wedged native call must not block process exit
# (the CI runner hang, 2026-08-12: 6 tests passed in 4s, then
# the file-timeout SIGKILL'd a process stuck in this lane).
return _run_bounded_on_exit_thread(
lambda: context.run(invoke), timeout
)
try:
return future.result(timeout=timeout)
except FuturesTimeoutError as exc:
raise TimeoutError(
f"Relay scope operation exceeded {timeout}s "
f"(session={session.session_id}); abandoning the native call "
"so the agent can continue — the span for this scope is lost"
) from exc
async def run_in_session_async(
self,
session: RelaySession,
callback: Callable[..., Any],
*args: Any,
allow_closing: bool = False,
**kwargs: Any,
) -> Any:
"""Create and await an operation inside the session's saved context."""
with session.lock:
if session.closing and not allow_closing:
raise RuntimeError("Hermes Relay session is closing")
if session.context is None or session.handle is None:
raise RuntimeError("Hermes Relay session context is unavailable")
relay_context = session.context.copy()
context = contextvars.copy_context()
for variable, value in relay_context.items():
context.run(variable.set, value)
async def invoke() -> Any:
self.relay.get_scope_stack()
result = callback(*args, **kwargs)
if inspect.isawaitable(result):
return await result
return result
task = context.run(asyncio.create_task, invoke())
return await task
def emit_mark(
self,
name: str,
event: dict[str, Any],
*,
data: Any = None,
metadata: Any = None,
) -> bool:
"""Emit a mark parented to the Hermes session identified by ``event``."""
session = self.ensure_session(event)
if session is None:
return False
self.run_in_session(
session,
self.relay.scope.event,
name,
handle=session.handle,
data=data,
metadata=metadata,
)
return True
def apply_tool_request_intercepts(
self,
*,
session_id: str,
tool_name: str,
args: dict[str, Any],
) -> dict[str, Any]:
"""Apply Relay request rewriting before Hermes authorizes a tool call."""
if not self.managed_execution_enabled():
return args
request_intercepts = getattr(
getattr(self.relay, "tools", None),
"request_intercepts",
None,
)
if not callable(request_intercepts):
return args
session = self.ensure_session({"session_id": session_id})
if session is None:
return args
result = self.run_in_session(
session,
request_intercepts,
tool_name,
args,
)
return result if isinstance(result, dict) else args
def close_session(self, event: dict[str, Any]) -> None:
"""Close one session scope and remove it from the core registry."""
session_id = _session_id(event)
with self._sessions_lock:
session = self._sessions.get(session_id)
if session is None:
with self._sessions_lock:
self._subagent_parents.pop(session_id, None)
self._subagent_parent_handles.pop(session_id, None)
return
failures: list[str] = []
with session.lock:
if session.closing:
return
session.closing = True
if session.handle is not None:
try:
self.run_in_session(
session,
self.relay.scope.pop,
session.handle,
output={},
metadata={
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: self.runtime_id,
},
allow_closing=True,
timeout=_SCOPE_OP_TIMEOUT,
)
except Exception as exc:
failures.append(f"session scope close failed: {exc}")
try:
try:
_scope_op_executor().submit(
self.relay.subscribers.flush
).result(timeout=_SCOPE_OP_TIMEOUT)
except RuntimeError:
# Interpreter shutdown: executor refuses new futures; flush
# on a bounded exit thread so a wedged pipeline cannot
# block process exit.
_run_bounded_on_exit_thread(
self.relay.subscribers.flush, _SCOPE_OP_TIMEOUT
)
except Exception as exc:
failures.append(f"subscriber flush failed: {exc}")
with self._sessions_lock:
if self._sessions.get(session_id) is session:
self._sessions.pop(session_id, None)
self._subagent_parents.pop(session_id, None)
self._subagent_parent_handles.pop(session_id, None)
if failures:
logger.warning(
"Hermes Relay session %s closed with errors: %s",
session_id,
"; ".join(failures),
)
def shutdown(self) -> None:
"""Close all core-owned Relay session scopes."""
with self._sessions_lock:
session_ids = list(self._sessions)
for session_id in session_ids:
self._safe(self.close_session, {"session_id": session_id})
if self._shutdown_registered:
try:
atexit.unregister(self.shutdown)
except Exception:
pass
self._shutdown_registered = False
@staticmethod
def _safe(callback: Callable[..., Any], *args: Any, **kwargs: Any) -> Any:
try:
return callback(*args, **kwargs)
except Exception:
logger.warning("Hermes Relay runtime operation failed", exc_info=True)
return None
@dataclass(frozen=True)
class NoopRelayRuntime:
"""Explicit reduced-capability host for platforms without Relay wheels."""
profile_key: str
reason: str
@property
def available(self) -> bool:
return False
def apply_tool_request_intercepts(
self,
*,
session_id: str,
tool_name: str,
args: dict[str, Any],
) -> dict[str, Any]:
del session_id, tool_name
return args
@staticmethod
def retain_managed_execution(consumer: str) -> None:
del consumer
@staticmethod
def release_managed_execution(consumer: str) -> None:
del consumer
@staticmethod
def managed_execution_enabled() -> bool:
return False
def shutdown(self) -> None:
"""No resources are allocated on unsupported platforms."""
RelayHost = RelayRuntime | NoopRelayRuntime
class RelayHostRegistry:
"""Own exactly one Relay host for each canonical Hermes profile."""
def __init__(self) -> None:
self._lock = threading.RLock()
self._hosts: dict[str, RelayHost] = {}
def for_profile(
self,
profile_key: str | None = None,
*,
create: bool = True,
) -> RelayHost | None:
key = profile_key or current_profile_key()
host = self._hosts.get(key)
if host is not None or not create:
return host
with self._lock:
host = self._hosts.get(key)
if host is not None or not create:
return host
try:
host = RelayRuntime(profile_key=key)
except Exception as exc:
logger.warning(
"Hermes Relay runtime initialization failed", exc_info=True
)
host = NoopRelayRuntime(profile_key=key, reason=str(exc))
self._hosts[key] = host
return host
def shutdown_profile(self, profile_key: str) -> None:
with self._lock:
host = self._hosts.pop(profile_key, None)
if host is not None:
host.shutdown()
def shutdown_all(self) -> None:
with self._lock:
hosts = list(self._hosts.values())
self._hosts.clear()
for host in hosts:
host.shutdown()
HOST_REGISTRY = RelayHostRegistry()
@dataclass
class ConversationLease:
"""A resumable reference to one profile-scoped conversation scope."""
profile_key: str
session_id: str
platform: str
host: RelayHost
session: RelaySession | None
parent_session_id: str = ""
released: bool = False
@dataclass
class RelayTurnContext:
"""Runtime-only context for one Hermes turn or top-level task."""
lease: ConversationLease
turn_id: str
task_id: str
handle: Any = None
logical_llm_calls: dict[str, Any] = field(default_factory=dict, repr=False)
logical_llm_lock: threading.RLock = field(
default_factory=threading.RLock,
repr=False,
)
finalize_lock: threading.RLock = field(
default_factory=threading.RLock,
repr=False,
)
_previous_turn: RelayTurnContext | None = field(default=None, repr=False)
_active_registered: bool = field(default=False, repr=False)
relay_enabled: bool = True
closed: bool = False
_CURRENT_TURN: contextvars.ContextVar[RelayTurnContext | None] = contextvars.ContextVar(
"hermes_relay_turn", default=None
)
# Depth of managed Relay callbacks executing on the current logical call path.
# Set >0 while the native Relay pipeline is mid-dispatch of a Hermes callback
# (tool or LLM). Nested managed execution inside that window is structurally
# broken — the native pipeline binds its Futures to the outer, blocked event
# loop — so resolve_execution_context() bypasses Relay while the flag is set.
# ContextVar so the marker follows contextvars.copy_context() into the worker
# threads / per-thread loops that tools use for their internal async work.
_MANAGED_CALLBACK_DEPTH: contextvars.ContextVar[int] = contextvars.ContextVar(
"hermes_relay_managed_callback_depth", default=0
)
class managed_callback_guard:
"""Mark the current context as inside a managed Relay callback.
Synchronous context manager used by the relay adapters around the
``invoke()`` callbacks they hand to the native pipeline. Everything the
callback transitively calls (including work it forwards to worker threads
via ``contextvars.copy_context()``) sees the marker and runs unmanaged.
"""
def __enter__(self) -> "managed_callback_guard":
self._token = _MANAGED_CALLBACK_DEPTH.set(_MANAGED_CALLBACK_DEPTH.get() + 1)
return self
def __exit__(self, *exc_info: Any) -> None:
_MANAGED_CALLBACK_DEPTH.reset(self._token)
class RelaySessionCoordinator:
"""Own semantic conversation and turn lifetimes for Hermes core."""
def __init__(self, registry: RelayHostRegistry = HOST_REGISTRY) -> None:
self.registry = registry
self._initializer_lock = threading.RLock()
self._session_initializers: dict[
str,
Callable[[RelayRuntime, dict[str, Any]], None],
] = {}
self._active_turns_lock = threading.RLock()
self._active_turns: dict[tuple[str, str], set[int]] = {}
def register_session_initializer(
self,
name: str,
callback: Callable[[RelayRuntime, dict[str, Any]], None],
) -> None:
"""Register idempotent profile/session preparation before scope creation."""
with self._initializer_lock:
self._session_initializers[name] = callback
def unregister_session_initializer(self, name: str) -> None:
"""Remove a previously registered session initializer."""
with self._initializer_lock:
self._session_initializers.pop(name, None)
def _prepare_session(
self,
host: RelayRuntime,
context: dict[str, Any],
) -> None:
with self._initializer_lock:
initializers = list(self._session_initializers.items())
for name, callback in initializers:
try:
callback(host, context)
except Exception:
logger.warning(
"Hermes Relay session initializer failed: %s",
name,
exc_info=True,
)
def acquire_conversation(
self,
*,
profile_key: str,
session_id: str,
platform: str,
parent_session_id: str = "",
model: str = "",
) -> ConversationLease:
host = self.registry.for_profile(profile_key)
if host is None:
host = NoopRelayRuntime(profile_key, "Relay host creation was disabled")
session = None
if isinstance(host, RelayRuntime):
try:
session_context = {
"profile_key": profile_key,
"session_id": session_id,
"platform": platform,
"parent_session_id": parent_session_id,
"model": model,
}
self._prepare_session(host, session_context)
metadata = {"hermes.execution_surface": platform or "unknown"}
if parent_session_id and parent_session_id != session_id:
session = host.register_subagent(
{
"parent_session_id": parent_session_id,
"child_session_id": session_id,
},
metadata=metadata,
)
else:
session = host.ensure_session(
{"session_id": session_id},
metadata=metadata,
)
except Exception:
logger.warning(
"Hermes Relay conversation initialization failed",
exc_info=True,
)
return ConversationLease(
profile_key=profile_key,
session_id=session_id,
platform=platform,
host=host,
session=session,
parent_session_id=parent_session_id,
)
def begin_turn(
self,
lease: ConversationLease,
*,
turn_id: str,
task_id: str,
) -> RelayTurnContext:
if lease.released:
raise RuntimeError("Hermes Relay conversation lease is released")
turn = RelayTurnContext(lease=lease, turn_id=turn_id, task_id=task_id)
key = (lease.profile_key, lease.session_id)
with self._active_turns_lock:
active = self._active_turns.get(key)
if active:
# A Relay session owns one physical scope stack. Concurrent
# Hermes turns would create sibling scopes on that stack, but
# their completion order is not guaranteed to be LIFO.
turn.relay_enabled = False
logger.warning(
"Skipping Relay instrumentation for concurrent Hermes turn "
"%s in session %s",
turn_id,
lease.session_id,
)
else:
self._active_turns[key] = {id(turn)}
turn._active_registered = True
if (
turn.relay_enabled
and isinstance(lease.host, RelayRuntime)
and lease.session is not None
):
# Session-span segmentation: consume a pending rotation (set by
# compaction) or the max_turns cap HERE — the only point where
# no turn scope is live on this session's stack, so the session
# scope can close/reopen without violating LIFO order.
try:
config = _segments_config()
session = lease.session
cap = config["max_turns"]
if (config["on_compaction"] and session.rotate_pending) or (
cap > 0 and session.segment_turns >= cap
):
reason = (
"compaction"
if config["on_compaction"] and session.rotate_pending
else "max_turns"
)
lease.host.rotate_session_scope(session, reason=reason)
except Exception:
logger.warning(
"Hermes Relay segment rotation failed", exc_info=True
)
try:
turn.handle = lease.host.run_in_session(
lease.session,
lease.host.relay.scope.push,
TURN_SCOPE,
lease.host.relay.ScopeType.Function,
handle=lease.session.handle,
input={},
metadata={
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: lease.host.runtime_id,
"hermes.execution_surface": lease.platform or "unknown",
},
timeout=_SCOPE_OP_TIMEOUT,
)
except Exception:
logger.warning("Hermes Relay turn initialization failed", exc_info=True)
turn._previous_turn = _CURRENT_TURN.get()
_CURRENT_TURN.set(turn)
return turn
def end_turn(
self,
turn: RelayTurnContext,
*,
outcome: str,
) -> None:
with turn.finalize_lock:
if turn.closed:
self._reset_turn_context(turn)
return
turn.closed = True
lease = turn.lease
try:
if isinstance(lease.host, RelayRuntime) and lease.session is not None:
self._finish_logical_calls(turn, outcome=outcome)
if turn.handle is not None:
try:
lease.host.run_in_session(
lease.session,
lease.host.relay.scope.pop,
turn.handle,
output={"outcome": outcome},
metadata={
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: lease.host.runtime_id,
},
timeout=_SCOPE_OP_TIMEOUT,
)
except Exception:
logger.warning(
"Hermes Relay turn finalization failed", exc_info=True
)
finally:
try:
# Segment turn accounting (max_turns rotation trigger).
if (
turn._active_registered
and isinstance(lease.host, RelayRuntime)
and lease.session is not None
):
with lease.session.lock:
lease.session.segment_turns += 1
except Exception: # noqa: BLE001 - accounting must never block
pass
try:
# Delegated agents own one turn. Close their conversation
# while the active-turn guard is still held so a parent
# timeout fallback cannot race this terminal boundary.
if (
lease.parent_session_id
and isinstance(lease.host, RelayRuntime)
):
lease.host.unregister_subagent({
"child_session_id": lease.session_id
})
except Exception:
logger.warning(
"Hermes Relay child conversation finalization failed",
exc_info=True,
)
finally:
self._unregister_active_turn(turn)
self._reset_turn_context(turn)
self._consume_deferred_close(lease)
def _consume_deferred_close(self, lease: Any) -> None:
"""Close a session whose rotating-compaction close was deferred.
``notify_session_compacted`` sets ``close_pending`` instead of
closing when the old session still has a live turn (closing then
would pop the session scope under the live turn scope — LIFO
violation). The turn that was live consumes the flag here, after
its own turn scope popped and it unregistered from the
active-turn table. Skips when another turn is still live on the
same session; that turn's end_turn will consume it instead.
"""
try:
if not (
isinstance(lease.host, RelayRuntime) and lease.session is not None
):
return
session = lease.session
with session.lock:
pending = session.close_pending and not session.closing
if not pending:
return
if self.has_active_turn(
profile_key=lease.profile_key, session_id=lease.session_id
):
return
lease.host.close_session({"session_id": lease.session_id})
except Exception: # noqa: BLE001 - telemetry must never block end_turn
logger.warning(
"Hermes Relay deferred session close failed", exc_info=True
)
def notify_session_compacted(
self,
*,
profile_key: str,
session_id: str,
old_session_id: str = "",
) -> None:
"""React to a completed compaction, per compaction mode.
In-place compaction (``old_session_id`` empty or equal to
``session_id``): flag the session for segment rotation at its next
turn boundary. Never rotates immediately — a compaction can
complete while a turn is live, and rotation under a live turn
scope would violate the scope stack's LIFO order; ``begin_turn``
consumes the flag.
Legacy rotating compaction (``old_session_id`` differs): the next
turn acquires a fresh Relay session under the new id on its own,
but the OLD session's scope would stay open forever — an
unexported orphan. Close it now so the pre-compaction segment
exports.
Unknown sessions and disabled config are silent no-ops; this
method must never add work or failure modes to the compaction
critical path.
"""
try:
if not _segments_config()["on_compaction"]:
return
host = self.registry.for_profile(profile_key)
if not isinstance(host, RelayRuntime):
return
if old_session_id and old_session_id != session_id:
# Rotating compaction: export the orphaned pre-compaction
# session scope (close_session is already bounded). If a
# turn is still LIVE on the old session, closing now would
# pop the session scope under the live turn scope (LIFO
# violation) — defer to that turn's end_turn instead.
with host._sessions_lock:
old_session = host._sessions.get(old_session_id)
if old_session is not None and self.has_active_turn(
profile_key=profile_key, session_id=old_session_id
):
with old_session.lock:
if not old_session.closing:
old_session.close_pending = True
return
host.close_session({"session_id": old_session_id})
return
with host._sessions_lock:
session = host._sessions.get(session_id)
if session is None:
return
with session.lock:
if not session.closing:
session.rotate_pending = True
except Exception: # noqa: BLE001 - telemetry must never block compaction
logger.warning(
"Hermes Relay compaction notification failed", exc_info=True
)
def has_active_turn(self, *, profile_key: str, session_id: str) -> bool:
"""Return whether a turn is still running for one profile/session."""
key = (profile_key, session_id)
with self._active_turns_lock:
return bool(self._active_turns.get(key))
def _unregister_active_turn(self, turn: RelayTurnContext) -> None:
if not turn._active_registered:
return
key = (turn.lease.profile_key, turn.lease.session_id)
with self._active_turns_lock:
active = self._active_turns.get(key)
if active is not None:
active.discard(id(turn))
if not active:
self._active_turns.pop(key, None)
turn._active_registered = False
def _reset_active_turns_for_tests(self) -> None:
with self._active_turns_lock:
self._active_turns.clear()
def finish_logical_calls(
self,
turn: RelayTurnContext,
*,
outcome: str,
) -> None:
"""Close logical LLM children before sibling task aggregation scopes."""
with turn.finalize_lock:
if turn.closed:
return
self._finish_logical_calls(turn, outcome=outcome)
@staticmethod
def _finish_logical_calls(
turn: RelayTurnContext,
*,
outcome: str,
) -> None:
lease = turn.lease
if not isinstance(lease.host, RelayRuntime) or lease.session is None:
return
with turn.logical_llm_lock:
logical_calls = list(turn.logical_llm_calls.items())
turn.logical_llm_calls.clear()
for index in range(len(logical_calls) - 1, -1, -1):
request_id, logical_handle = logical_calls[index]
try:
lease.host.run_in_session(
lease.session,
lease.host.relay.scope.pop,
logical_handle,
output={"outcome": outcome},
metadata={
RUNTIME_SCHEMA_KEY: RUNTIME_SCHEMA_VERSION,
RUNTIME_INSTANCE_KEY: lease.host.runtime_id,
},
timeout=_SCOPE_OP_TIMEOUT,
)
except Exception:
with turn.logical_llm_lock:
# Relay scopes are stack-owned. If the newest remaining
# handle cannot close, older handles cannot close safely
# either, so retain the unclosed prefix for diagnostics.
for pending_request_id, pending_handle in logical_calls[
: index + 1
]:
turn.logical_llm_calls.setdefault(
pending_request_id,
pending_handle,
)
logger.warning(
"Hermes Relay logical LLM finalization failed",
exc_info=True,
)
break
@staticmethod
def _reset_turn_context(turn: RelayTurnContext) -> None:
"""Unwind ``turn`` without disturbing a newer context-local turn."""
if _CURRENT_TURN.get() is not turn:
return
previous = turn._previous_turn
seen = {id(turn)}
while previous is not None and previous.closed:
if id(previous) in seen:
previous = None
break
seen.add(id(previous))
previous = previous._previous_turn
_CURRENT_TURN.set(previous)
@staticmethod
def release_conversation(lease: ConversationLease) -> None:
"""Release a caller lease without closing a resumable conversation."""
lease.released = True
def finalize_conversation(
self,
*,
profile_key: str,
session_id: str,
) -> None:
host = self.registry.for_profile(profile_key, create=False)
if isinstance(host, RelayRuntime):
host.close_session({"session_id": session_id})
def shutdown_profile(self, profile_key: str) -> None:
self.registry.shutdown_profile(profile_key)
SESSION_COORDINATOR = RelaySessionCoordinator()
def current_turn() -> RelayTurnContext | None:
"""Return the turn context inherited by current async and thread work."""
return _CURRENT_TURN.get()
def relay_instrumentation_enabled() -> bool:
"""Return whether this inherited turn may create Relay instrumentation."""
turn = current_turn()
return turn is None or (turn.relay_enabled and not turn.closed)
def active_turn(session_id: str | None = None) -> RelayTurnContext | None:
"""Return a live turn only when it belongs to the active profile/session."""
turn = current_turn()
if (
turn is None
or not turn.relay_enabled
or turn.closed
or turn.lease.released
):
return None
if turn.lease.profile_key != current_profile_key():
return None
if session_id is not None and turn.lease.session_id != session_id:
return None
if isinstance(turn.lease.host, RelayRuntime):
if turn.lease.session is None:
return None
if turn.lease.host.get_session(turn.lease.session_id) is not turn.lease.session:
return None
return turn
def resolve_execution_context(
session_id: str,
) -> tuple[RelayRuntime | None, RelaySession | None, Any]:
"""Resolve one active turn/session parent for managed Relay execution."""
if _MANAGED_CALLBACK_DEPTH.get() > 0:
# A managed Relay callback is already executing on this logical call
# path (e.g. the native ``tools.execute`` pipeline is mid-dispatch of
# a Hermes tool). Nested managed execution here is structurally
# impossible: the native pipeline binds its Futures to the OUTER
# call's event loop, which is blocked inside the synchronous tool
# callback until the tool returns. A nested managed LLM call (the
# vision_analyze auxiliary path) therefore awaits a foreign-loop
# Future that can never complete — "attached to a different loop"
# at best, deadlock at worst, and "Event loop is closed" during
# shutdown when the orphaned Future is completed late (#77244).
# Run nested calls unmanaged; the outer tool scope still records
# the tool-level event for observability.
return None, None, None
inherited_turn = current_turn()
if inherited_turn is not None and (
not inherited_turn.relay_enabled or inherited_turn.closed
):
return None, None, None
turn = active_turn(session_id)
if (
turn is not None
and isinstance(turn.lease.host, RelayRuntime)
and turn.lease.session is not None
):
session = turn.lease.session
return turn.lease.host, session, turn.handle or session.handle
# Managed-execution consumers create and retain the profile host before
# reaching an out-of-turn adapter. Do not initialize Relay for the default
# no-consumer path.
runtime = get_runtime(create=False)
if runtime is None:
return None, None, None
if not runtime.managed_execution_enabled():
return None, None, None
session = runtime.get_session(session_id)
if session is None:
session = runtime.ensure_session({"session_id": session_id})
return runtime, session, None if session is None else session.handle
def emit_mark(
name: str,
*,
session_id: str,
data: Any = None,
metadata: Any = None,
) -> bool:
"""Emit a fail-open Relay mark under a Hermes session."""
runtime = get_runtime(create=False)
if runtime is None:
return False
try:
return runtime.emit_mark(
name,
{"session_id": session_id},
data=data,
metadata=metadata,
)
except Exception:
logger.warning("Hermes Relay mark failed: %s", name, exc_info=True)
return False
def apply_tool_request_intercepts(
*,
session_id: str,
tool_name: str,
args: dict[str, Any],
) -> dict[str, Any]:
"""Return Relay-rewritten arguments at Hermes's authorization boundary."""
if not session_id:
return args
runtime = get_runtime(create=False)
if runtime is None:
return args
return runtime.apply_tool_request_intercepts(
session_id=session_id,
tool_name=tool_name,
args=args,
)
def ensure_session(*, session_id: str, **context: Any) -> RelaySession | None:
"""Create or return the shared Relay session used by Hermes core."""
runtime = get_runtime()
if runtime is None:
return None
try:
return runtime.ensure_session({"session_id": session_id, **context})
except Exception:
logger.warning("Hermes Relay session initialization failed", exc_info=True)
return None
def run_in_session(
session_id: str,
callback: Callable[..., Any],
*args: Any,
**kwargs: Any,
) -> Any:
"""Run a scope, LLM, or tool API against a shared Hermes session."""
runtime = get_runtime()
if runtime is None:
raise RuntimeError("Hermes Relay runtime is unavailable")
session = runtime.get_session(session_id)
if session is None:
session = runtime.ensure_session({"session_id": session_id})
if session is None:
raise RuntimeError("Hermes Relay session is unavailable")
return runtime.run_in_session(session, callback, *args, **kwargs)
async def run_in_session_async(
session_id: str,
callback: Callable[..., Any],
*args: Any,
**kwargs: Any,
) -> Any:
"""Await a Relay operation inside a shared Hermes session context."""
runtime = get_runtime()
if runtime is None:
raise RuntimeError("Hermes Relay runtime is unavailable")
session = runtime.get_session(session_id)
if session is None:
session = runtime.ensure_session({"session_id": session_id})
if session is None:
raise RuntimeError("Hermes Relay session is unavailable")
return await runtime.run_in_session_async(session, callback, *args, **kwargs)
def get_session_handle(session_id: str) -> Any:
"""Return the shared Relay handle for direct core instrumentation."""
runtime = get_runtime(create=False)
return None if runtime is None else runtime.get_session_handle(session_id)
def _is_relay_wrapped_callback_error(
relay_error: BaseException,
callback_error: BaseException,
) -> bool:
"""Match Relay's native callback wrapper without masking policy errors."""
if relay_error is callback_error:
return True
if not isinstance(relay_error, RuntimeError):
return False
callback_type = callback_error.__class__
type_names = {
callback_type.__name__,
callback_type.__qualname__,
f"{callback_type.__module__}.{callback_type.__qualname__}",
}
message = str(relay_error)
return any(
message.startswith(f"internal error: {type_name}: {callback_error}")
for type_name in type_names
)
def get_runtime(
*,
create: bool = True,
profile_key: str | None = None,
) -> RelayRuntime | None:
"""Return the Relay host for the active Hermes profile."""
host = HOST_REGISTRY.for_profile(profile_key, create=create)
return host if isinstance(host, RelayRuntime) else None
def get_host(
*,
create: bool = True,
profile_key: str | None = None,
) -> RelayHost | None:
"""Return the explicit real or reduced-capability host for a profile."""
return HOST_REGISTRY.for_profile(profile_key, create=create)
def current_profile_key() -> str:
"""Return the canonical profile identity used for runtime isolation."""
home = get_hermes_home().expanduser()
if not home.is_absolute():
return str(home.resolve())
raw = str(home)
cached = _PROFILE_KEY_CACHE.get(raw)
if cached is not None:
return cached
resolved = str(home.resolve())
return _PROFILE_KEY_CACHE.setdefault(raw, resolved)
def _load_nemo_relay() -> Any:
"""Load the binding only when a producer or consumer needs Relay."""
return importlib.import_module("nemo_relay")
def _session_id(event: dict[str, Any]) -> str:
return str(event.get("session_id") or "")
def _reset_for_tests() -> None:
"""Reset all profile-scoped Relay hosts for isolated tests."""
SESSION_COORDINATOR._reset_active_turns_for_tests()
HOST_REGISTRY.shutdown_all()
_PROFILE_KEY_CACHE.clear()