0.13.0 Dynamic Group stream: missing run_id in agent.py + BlockManager.get_block_by_id attr error (async API)

[Kyujaq]

Version: letta/letta:0.13.0 (Docker)

Endpoint(s):

POST /v1/groups//messages and /messages/stream

Behavior:

SSE opens (200 + text/event-stream), then error

Errors observed:

TypeError: convert_message_creates_to_messages() missing 1 required positional argument: 'run_id' called from /app/letta/agent.py:783

AttributeError: 'BlockManager' object has no attribute 'get_block_by_id' (seems new async API is get_block_by_id_async)

Notes:

Agents stream fine via /v1/agents//messages/stream

We fixed persona/human blocks + optional composio imports; only group path fails

Expected: group step/step_stream should emit events

Actual: internal error payload {"error":"Stream failed (internal error occurred)"} then [DONE]

File "/app/letta/agent.py", line 897, in inner_step
blocks=[self.block_manager.get_block_by_id(block.id, actor=self.user) for block in self.agent_state.memory.get_blocks()],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 897, in
blocks=[self.block_manager.get_block_by_id(block.id, actor=self.user) for block in self.agent_state.memory.get_blocks()],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'BlockManager' object has no attribute 'get_block_by_id'
Letta.letta.utils - ERROR - multi_agent_step_thread failed with AttributeError: 'BlockManager' object has no attribute 'get_block_by_id'
Traceback (most recent call last):
File "/app/letta/utils.py", line 1115, in wrapper
await coro
File "/usr/lib/python3.11/asyncio/threads.py", line 25, in to_thread
return await loop.run_in_executor(None, func_call)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.11/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/groups/round_robin_multi_agent.py", line 163, in step
raise e
File "/app/letta/groups/round_robin_multi_agent.py", line 109, in step
usage_stats = participant_agent.step(
^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/otel/tracing.py", line 273, in sync_wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 795, in step
step_response = self.inner_step(
^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 1088, in inner_step

Letta.letta.server.rest_api.utils - ERROR - Expected LettaUsageStatistics, got <class 'NoneType'>
Letta.letta.server.rest_api.utils - ERROR - Caught unexpected Exception: Expected LettaUsageStatistics, got <class 'NoneType'>
Letta.letta.llm_api.openai_client - WARNING - Failed to convert tool function to structured output, tool=type='function' function=FunctionSchema(name='ha_call_service', description='Call a Home Assistant service synchronously.', parameters={'type': 'object', 'properties': {'thinking': {'type': 'string', 'description': 'Deep inner monologue private to you only.'}, 'domain': {'type': 'string', 'description': "HA domain (e.g. 'light')."}, 'service': {'type': 'string', 'description': "Service name (e.g. 'turn_on')."}, 'data': {'type': 'object', 'description': 'Optional service payload.'}, 'request_heartbeat': {'type': 'boolean', 'description': 'Request an immediate heartbeat after function execution. You MUST set this value to True if you want to send a follow-up message or run a follow-up tool call (chain multiple tools together). If set to False (the default), then the chain of execution will end immediately after this function call.'}}, 'required': ['thinking', 'domain', 'service', 'request_heartbeat'], 'additionalProperties': False}, strict=True), error=Property {'type': 'object', 'description': 'Optional service payload.'} of type object is missing properties
Letta.letta.llm_api.openai_client - WARNING - Failed to convert tool function to structured output, tool=type='function' function=FunctionSchema(name='alert_notify', description='Send a notification payload to the configured webhook.', parameters={'type': 'object', 'properties': {'thinking': {'type': 'string', 'description': 'Deep inner monologue private to you only.'}, 'message': {'type': 'string', 'description': 'Human readable message.'}, 'severity': {'type': 'string', 'description': 'Severity label (info|warn|critical).'}, 'data': {'type': 'object', 'description': 'Optional structured payload.'}, 'request_heartbeat': {'type': 'boolean', 'description': 'Request an immediate heartbeat after function execution. You MUST set this value to True if you want to send a follow-up message or run a follow-up tool call (chain multiple tools together). If set to False (the default), then the chain of execution will end immediately after this function call.'}}, 'required': ['thinking', 'message', 'request_heartbeat'], 'additionalProperties': False}, strict=True), error=Property {'type': 'object', 'description': 'Optional structured payload.'} of type object is missing properties
Letta.letta.llm_api.openai_client - WARNING - Failed to convert tool function to structured output, tool=type='function' function=FunctionSchema(name='memory_write', description='Persist text into the memory embedding store.', parameters={'type': 'object', 'properties': {'thinking': {'type': 'string', 'description': 'Deep inner monologue private to you only.'}, 'text': {'type': 'string', 'description': 'Content to store.'}, 'kind': {'type': 'string', 'description': 'Memory kind label.'}, 'source': {'type': 'string', 'description': 'Logical source label.'}, 'meta': {'type': 'object', 'description': 'Optional metadata dict.'}, 'memory_id': {'type': 'string', 'description': 'Optional existing identifier for upsert.'}, 'request_heartbeat': {'type': 'boolean', 'description': 'Request an immediate heartbeat after function execution. You MUST set this value to True if you want to send a follow-up message or run a follow-up tool call (chain multiple tools together). If set to False (the default), then the chain of execution will end immediately after this function call.'}}, 'required': ['thinking', 'text', 'request_heartbeat'], 'additionalProperties': False}, strict=True), error=Property {'type': 'object', 'description': 'Optional metadata dict.'} of type object is missing properties
Letta.letta.llm_api.openai_client - WARNING - Failed to convert tool function to structured output, tool=type='function' function=FunctionSchema(name='memory_update', description='Update an existing memory entry (same contract as memory_write).', parameters={'type': 'object', 'properties': {'thinking': {'type': 'string', 'description': 'Deep inner monologue private to you only.'}, 'memory_id': {'type': 'string', 'description': 'Identifier returned by memory_write that should be updated.'}, 'text': {'type': 'string', 'description': 'New content that replaces the previous memory text.'}, 'kind': {'type': 'string', 'description': "Optional classification label for the memory (defaults to 'note')."}, 'source': {'type': 'string', 'description': "Optional logical source label for auditing (defaults to 'agent')."}, 'meta': {'type': 'object', 'description': 'Optional metadata dictionary stored alongside the memory.'}, 'request_heartbeat': {'type': 'boolean', 'description': 'Request an immediate heartbeat after function execution. You MUST set this value to True if you want to send a follow-up message or run a follow-up tool call (chain multiple tools together). If set to False (the default), then the chain of execution will end immediately after this function call.'}}, 'required': ['thinking', 'memory_id', 'text', 'request_heartbeat'], 'additionalProperties': False}, strict=True), error=Property {'type': 'object', 'description': 'Optional metadata dictionary stored alongside the memory.'} of type object is missing properties
httpx - INFO - HTTP Request: POST http://ollama-text:11434/v1/chat/completions "HTTP/1.1 200 OK"
Letta.letta.llm_api.helpers - WARNING - Overwriting existing inner monologue () with kwarg (User has logged in for the first time. Let's start off with a friendly greeting.)
INFO: 172.21.0.1:59710 - "POST /v1/agents/agent-51220cc3-4648-4096-9bf2-c2484ef15571/messages/stream HTTP/1.1" 200 OK
Letta.agent-51220cc3-4648-4096-9bf2-c2484ef15571 - INFO - Running final update. Step Progression: StepProgression.FINISHED
Letta.agent-51220cc3-4648-4096-9bf2-c2484ef15571 - INFO - Total tokens 3678 does not exceed configured max tokens 8192, passing summarizing w/o force.
Letta.letta.services.run_manager - WARNING - Run run-d1257c86-1d84-485c-b81b-db88e978ee70 completed without a completed_at timestamp
INFO: 172.21.0.1:45288 - "POST /v1/groups/group-ef05bb2b-fc6e-4665-8448-0f2acb229573/messages/stream HTTP/1.1" 200 OK
Letta.letta.agent - ERROR - step() failed
messages = [Message(created_by_id=None, last_updated_by_id=None, created_at=datetime.datetime(2025, 10, 25, 22, 16, 34, 658066, tzinfo=datetime.timezone.utc), updated_at=None, id='message-f328e4b6-a7bf-4117-9a41-0dee8ea49cb0', agent_id='agent-51220cc3-4648-4096-9bf2-c2484ef15571', model=None, role=<MessageRole.user: 'user'>, content=[TextContent(type=<MessageContentType.text: 'text'>, text='{\n "type": "user_message",\n "message": "Say hello in 3 words",\n "time": "2025-10-25 10:16:34 PM UTC+0000"\n}', signature=None)], name=None, tool_calls=None, tool_call_id=None, step_id=None, run_id='d85ed853-5aaa-4777-8058-815ea18593af', otid=None, tool_returns=None, group_id='group-ef05bb2b-fc6e-4665-8448-0f2acb229573', sender_id=None, batch_item_id=None, is_err=None, approval_request_id=None, approve=None, denial_reason=None, approvals=None)]
error = 'BlockManager' object has no attribute 'get_block_by_id'
Letta.letta.agent - ERROR - step() failed with an unrecognized exception: ''BlockManager' object has no attribute 'get_block_by_id''
Traceback (most recent call last):
File "/app/letta/agent.py", line 897, in inner_step
blocks=[self.block_manager.get_block_by_id(block.id, actor=self.user) for block in self.agent_state.memory.get_blocks()],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 897, in
blocks=[self.block_manager.get_block_by_id(block.id, actor=self.user) for block in self.agent_state.memory.get_blocks()],
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'BlockManager' object has no attribute 'get_block_by_id'
Letta.letta.utils - ERROR - multi_agent_step_thread failed with AttributeError: 'BlockManager' object has no attribute 'get_block_by_id'
Traceback (most recent call last):
File "/app/letta/utils.py", line 1115, in wrapper
await coro
File "/usr/lib/python3.11/asyncio/threads.py", line 25, in to_thread
return await loop.run_in_executor(None, func_call)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.11/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/groups/round_robin_multi_agent.py", line 163, in step
raise e
File "/app/letta/groups/round_robin_multi_agent.py", line 109, in step
usage_stats = participant_agent.step(
^^^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/otel/tracing.py", line 273, in sync_wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 795, in step
step_response = self.inner_step(
^^^^^^^^^^^^^^^^
File "/app/letta/agent.py", line 1088, in inner_step

Letta.letta.server.rest_api.utils - ERROR - Expected LettaUsageStatistics, got <class 'NoneType'>
Letta.letta.server.rest_api.utils - ERROR - Caught unexpected Exception: Expected LettaUsageStatistics, got <class 'NoneType'>