opengpts
opengpts copied to clipboard
langchain-ai
docker compose failing - frontend connecting to backend?
Having issues running via docker compose, screen comes up "half blank" and + New Bot does not work etc. Logs suggest issues connecting opengpts-frontend to backend?
Populated .env simply with an OPENAI_API_KEY="sk-...
Docker Version: 24.0.7 (Mac 14.4)
docker compose up
Containers all start, Uvicorn serving the "1/2 blank" page and opengpts-backend container generating errors, However the key issue may well be the opengpts-frontend see log output below
opengpts-frontend |
opengpts-frontend | VITE v4.5.0 ready in 117 ms
opengpts-frontend |
opengpts-frontend | ➜ Local: http://localhost:5173/
opengpts-frontend | ➜ Network: http://172.21.0.3:5173/
opengpts-backend | Process SpawnProcess-1:
opengpts-backend | Traceback (most recent call last):
opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 314, in _bootstrap
opengpts-backend | self.run()
opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 108, in run
opengpts-backend | self._target(*self._args, **self._kwargs)
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
opengpts-backend | target(sockets=sockets)
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 61, in run
opengpts-backend | return asyncio.run(self.serve(sockets=sockets))
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 190, in run
opengpts-backend | return runner.run(main)
opengpts-backend | ^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 118, in run
opengpts-backend | return self._loop.run_until_complete(task)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete
opengpts-backend | return future.result()
opengpts-backend | ^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 68, in serve
opengpts-backend | config.load()
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/config.py", line 467, in load
opengpts-backend | self.loaded_app = import_from_string(self.app)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/importer.py", line 21, in import_from_string
opengpts-backend | module = importlib.import_module(module_str)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
opengpts-backend | return _bootstrap._gcd_import(name[level:], package, level)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
opengpts-backend | File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
opengpts-backend | File "<frozen importlib._bootstrap>", line 1147, in _find_and_load_unlocked
opengpts-backend | File "<frozen importlib._bootstrap>", line 690, in _load_unlocked
opengpts-backend | File "<frozen importlib._bootstrap_external>", line 940, in exec_module
opengpts-backend | File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
opengpts-backend | File "/backend/app/server.py", line 9, in <module>
opengpts-backend | from app.api import router as api_router
opengpts-backend | File "/backend/app/api/__init__.py", line 3, in <module>
opengpts-backend | from app.api.assistants import router as assistants_router
opengpts-backend | File "/backend/app/api/assistants.py", line 7, in <module>
opengpts-backend | import app.storage as storage
opengpts-backend | File "/backend/app/storage.py", line 9, in <module>
opengpts-backend | from app.agent import AgentType, get_agent_executor
opengpts-backend | File "/backend/app/agent.py", line 23, in <module>
opengpts-backend | from app.tools import (
opengpts-backend | File "/backend/app/tools.py", line 163, in <module>
opengpts-backend | v()
opengpts-backend | File "/backend/app/tools.py", line 124, in _get_action_server
opengpts-backend | toolkit = ActionServerToolkit(
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/pydantic/main.py", line 341, in __init__
opengpts-backend | raise validation_error
opengpts-backend | pydantic.error_wrappers.ValidationError: 2 validation errors for ActionServerToolkit
opengpts-backend | url
opengpts-backend | none is not an allowed value (type=type_error.none.not_allowed)
opengpts-backend | api_key
opengpts-backend | none is not an allowed value (type=type_error.none.not_allowed)
opengpts-frontend | Error: connect ECONNREFUSED 172.21.0.4:8000
opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
opengpts-frontend | 11:37:25 AM [vite] http proxy error at /runs/config_schema:
opengpts-frontend | Error: connect ECONNREFUSED 172.21.0.4:8000
opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
opengpts-frontend | 11:37:25 AM [vite] http proxy error at /assistants/public/:
opengpts-frontend | Error: connect ECONNREFUSED 172.21.0.4:8000
opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
opengpts-frontend | 11:37:25 AM [vite] http proxy error at /assistants/:
opengpts-frontend | Error: connect ECONNREFUSED 172.21.0.4:8000
I figured out that this happens because of missing API key for robocorp agent tool, I tried adding the keys, and it moves forward, but now it is stuck because it cannot find openai api schema on the endpoint that I specify for robocorp agent tool. I am trying to figure out how to get it to stop expecting to use robocorp, that should ideally solve the issue. Perhaps. See the trace below: opengpts-redis | 8:M 03 Feb 2024 14:23:34.604 * Server initialized
opengpts-redis | 8:M 03 Feb 2024 14:23:34.605 * Ready to accept connections tcp
opengpts-frontend | yarn run v1.22.19
opengpts-frontend | $ vite --host
opengpts-backend | INFO: Will watch for changes in these directories: ['/backend']
opengpts-backend | INFO: Uvicorn running on http://localhost:8000 (Press CTRL+C to quit)
opengpts-backend | INFO: Started reloader process [1] using StatReload
opengpts-frontend |
opengpts-frontend | VITE v4.5.0 ready in 106 ms
opengpts-frontend |
opengpts-frontend | ➜ Local: http://localhost:5173/
opengpts-frontend | ➜ Network: http://xx:5173/
opengpts-backend | Process SpawnProcess-1:
opengpts-backend | Traceback (most recent call last):
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/langchain_robocorp/toolkits.py", line 121, in get_tools
opengpts-backend | api_spec = reduce_openapi_spec(self.url, json_spec)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/langchain_robocorp/_common.py", line 30, in reduce_openapi_spec
opengpts-backend | for route, operation in spec["paths"].items()
opengpts-backend | ~~~~^^^^^^^^^
opengpts-backend | KeyError: 'paths'
opengpts-backend |
opengpts-backend | During handling of the above exception, another exception occurred:
opengpts-backend |
opengpts-backend | Traceback (most recent call last):
opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 314, in _bootstrap
opengpts-backend | self.run()
opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 108, in run
opengpts-backend | self._target(*self._args, **self._kwargs)
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_started
opengpts-backend | target(sockets=sockets)
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 61, in run
opengpts-backend | return asyncio.run(self.serve(sockets=sockets))
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 190, in run
opengpts-backend | return runner.run(main)
opengpts-backend | ^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 118, in run
opengpts-backend | return self._loop.run_until_complete(task)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/asyncio/base_events.py", line 653, in run_until_complete
opengpts-backend | return future.result()
opengpts-backend | ^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 68, in serve
opengpts-backend | config.load()
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/config.py", line 467, in load
opengpts-backend | self.loaded_app = import_from_string(self.app)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/importer.py", line 21, in import_from_string
opengpts-backend | module = importlib.import_module(module_str)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "/usr/local/lib/python3.11/importlib/init.py", line 126, in import_module
opengpts-backend | return _bootstrap._gcd_import(name[level:], package, level)
opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
opengpts-backend | File "
Update: I got the bots working (RAG and simple chat) if I comment out, them in tools.py in backend. I also made a small change in the makefile. Still trying: but here is the change:
tools.py in Backend:
class AvailableTools(str, Enum):
# ACTION_SERVER = "Action Server by Robocorp"
# DDG_SEARCH = "DDG Search"
# TAVILY = "Search (Tavily)"
# TAVILY_ANSWER = "Search (short answer, Tavily)"
# RETRIEVAL = "Retrieval"
# ARXIV = "Arxiv"
# YOU_SEARCH = "You.com Search"
# SEC_FILINGS = "SEC Filings (Kay.ai)"
# PRESS_RELEASES = "Press Releases (Kay.ai)"
# PUBMED = "PubMed"
WIKIPEDIA = "Wikipedia"
TOOLS = {
# AvailableTools.ACTION_SERVER: _get_action_server,
# AvailableTools.DDG_SEARCH: _get_duck_duck_go,
# AvailableTools.ARXIV: _get_arxiv,
# AvailableTools.YOU_SEARCH: _get_you_search,
# AvailableTools.SEC_FILINGS: _get_sec_filings,
# AvailableTools.PRESS_RELEASES: _get_press_releases,
# AvailableTools.PUBMED: _get_pubmed,
# AvailableTools.TAVILY: _get_tavily,
AvailableTools.WIKIPEDIA: _get_wikipedia,
# AvailableTools.TAVILY_ANSWER: _get_tavily_answer,
}
TOOL_OPTIONS = {e.value: e.value for e in AvailableTools}
The dockerfile in backend changed IP address:
ENTRYPOINT [ "uvicorn", "app.server:app", "--host", "172.18.0.4" ]
at least the bots are working. I will continue to check what is going on. If someone finds out, pl let us know!
Thanks @abhinandanshahdev - so now I have progress by adding to .env (eg copying from .env-example)
ROBOCORP_ACTION_SERVER_URL=https://dummy-action-server.robocorp.link
ROBOCORP_ACTION_SERVER_KEY=dummy-api-key
However new issues remain. UI renders but on a save I get
opengpts-frontend | ➜ Network: http://172.21.0.3:5173/
opengpts-backend | No UI directory found, serving API only.
opengpts-backend | INFO: Started server process [8]
opengpts-backend | INFO: Waiting for application startup.
opengpts-backend | INFO: Application startup complete.
opengpts-backend | INFO: 172.21.0.3:44840 - "GET /runs/config_schema HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.3:44854 - "GET /threads/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.3:44864 - "GET /assistants/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.3:44880 - "GET /assistants/public/ HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.3:45142 - "PUT /assistants/407845c4-1d48-4915-8e6e-f25d9b4243a8 HTTP/1.1" 422 Unprocessable Entity
Try adding all the env variables that are present in .env.example, even if they are there as placeholders.
So you should have your OPENAI_API_KEY, and the rest you can use the same placeholders.
Try adding all the env variables that are present in
.env.example, even if they are there as placeholders. So you should have your OPENAI_API_KEY, and the rest you can use the same placeholders.
Tried this, just rebased, .env is a copy of .env.example except for a correct OPENAI_API_KEY, performed a docker-compose up --build and unfortunately, I'm still seeing this:
opengpts-frontend | ➜ Local: http://localhost:5173/
opengpts-frontend | ➜ Network: http://172.21.0.2:5173/
opengpts-backend | No UI directory found, serving API only.
opengpts-backend | INFO: Started server process [8]
opengpts-backend | INFO: Waiting for application startup.
opengpts-backend | INFO: Application startup complete.
opengpts-backend | INFO: 172.21.0.2:45438 - "GET /runs/config_schema HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.2:45444 - "GET /threads/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.2:45450 - "GET /assistants/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.2:45432 - "GET /assistants/public/ HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.2:55620 - "PUT /assistants/722cee00-53a6-4775-8dd7-c0b97853bff3 HTTP/1.1" 422 Unprocessable Entity
Which effectively means that the opening screen renders fine but I cannot create, or save Bots or Chats. When I have more time I'll dig deeper into these messages in particular:
opengpts-frontend | ➜ Local: http://localhost:5173/
opengpts-frontend | ➜ Network: http://172.21.0.2:5173/
opengpts-backend | No UI directory found, serving API only.
opengpts-backend | INFO: Started server process [8]
opengpts-backend | INFO: Waiting for application startup.
opengpts-backend | INFO: Application startup complete.
opengpts-backend | INFO: 172.21.0.2:45438 - "GET /runs/config_schema HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.2:45444 - "GET /threads/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.2:45450 - "GET /assistants/ HTTP/1.1" 422 Unprocessable Entity
opengpts-backend | INFO: 172.21.0.2:45432 - "GET /assistants/public/ HTTP/1.1" 200 OK
opengpts-backend | INFO: 172.21.0.2:55620 - "PUT /assistants/722cee00-53a6-4775-8dd7-c0b97853bff3 HTTP/1.1" 422 Unprocessable Entity
Same here, the project doesnt work. Very frustrating.
Using .env.example copy, only replacing OpenAI_API_KEY.
opengpts-redis | 8:M 24 Feb 2024 14:30:19.426 * Server initialized opengpts-redis | 8:M 24 Feb 2024 14:30:19.426 * Ready to accept connections tcp opengpts-frontend | yarn run v1.22.19 opengpts-frontend | $ vite --host opengpts-backend | INFO: Will watch for changes in these directories: ['/backend'] opengpts-backend | INFO: Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit) opengpts-backend | INFO: Started reloader process [1] using StatReload opengpts-frontend | opengpts-frontend | VITE v4.5.0 ready in 122 ms opengpts-frontend | opengpts-frontend | ➜ Local: http://localhost:5173/ opengpts-frontend | ➜ Network: http://172.19.0.3:5173/ opengpts-backend | Process SpawnProcess-1: opengpts-backend | Traceback (most recent call last): opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 314, in _bootstrap opengpts-backend | self.run() opengpts-backend | File "/usr/local/lib/python3.11/multiprocessing/process.py", line 108, in run opengpts-backend | self._target(*self._args, **self._kwargs) opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/_subprocess.py", line 76, in subprocess_start ed opengpts-backend | target(sockets=sockets) opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 61, in run opengpts-backend | return asyncio.run(self.serve(sockets=sockets)) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 190, in run opengpts-backend | return runner.run(main) opengpts-backend | ^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/asyncio/runners.py", line 118, in run opengpts-backend | return self._loop.run_until_complete(task) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/asyncio/base_events.py", line 654, in run_until_complete opengpts-backend | return future.result() opengpts-backend | ^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/server.py", line 68, in serve opengpts-backend | config.load() opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/config.py", line 467, in load opengpts-backend | self.loaded_app = import_from_string(self.app) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/site-packages/uvicorn/importer.py", line 21, in import_from_string opengpts-backend | module = importlib.import_module(module_str) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/importlib/init.py", line 126, in import_module opengpts-backend | return _bootstrap._gcd_import(name[level:], package, level) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "
", line 1204, in _gcd_import opengpts-backend | File " ", line 1176, in _find_and_load opengpts-backend | File " ", line 1147, in _find_and_load_unlocked opengpts-backend | File " ", line 690, in _load_unlocked opengpts-backend | File " ", line 940, in exec_module opengpts-backend | File " ", line 241, in _call_with_frames_removed opengpts-backend | File "/backend/app/server.py", line 9, in opengpts-backend | from app.api import router as api_router opengpts-backend | File "/backend/app/api/init.py", line 3, in opengpts-backend | from app.api.assistants import router as assistants_router opengpts-backend | File "/backend/app/api/assistants.py", line 7, in opengpts-backend | import app.storage as storage opengpts-backend | File "/backend/app/storage.py", line 10, in opengpts-backend | from app.agent import AgentType, get_agent_executor opengpts-backend | File "/backend/app/agent.py", line 200, in opengpts-backend | ConfigurableChatBot(llm=LLMType.GPT_35_TURBO, checkpoint=CHECKPOINTER) opengpts-backend | File "/backend/app/agent.py", line 189, in init opengpts-backend | chatbot = get_chatbot(llm, system_message) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/backend/app/agent.py", line 155, in get_chatbot opengpts-backend | llm = get_openai_llm() opengpts-backend | ^^^^^^^^^^^^^^^^ opengpts-backend | File "/backend/app/llms.py", line 14, in get_openai_llm opengpts-backend | http_client = httpx.AsyncClient(proxies=proxy_url) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/site-packages/httpx/_client.py", line 1395, in init opengpts-backend | proxy_map = self._get_proxy_map(proxies, allow_env_proxies) opengpts-backend | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/site-packages/httpx/_client.py", line 228, in _get_proxy_map opengpts-backend | proxy = Proxy(url=proxies) if isinstance(proxies, (str, URL)) else proxies opengpts-backend | ^^^^^^^^^^^^^^^^^^ opengpts-backend | File "/usr/local/lib/python3.11/site-packages/httpx/_config.py", line 334, in init opengpts-backend | raise ValueError(f"Unknown scheme for proxy URL {url!r}") opengpts-backend | ValueError: Unknown scheme for proxy URL URL('your_proxy_url') opengpts-frontend | 2:30:40 PM [vite] http proxy error at /threads/: opengpts-frontend | Error: connect ECONNREFUSED 172.19.0.4:8000 opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16) opengpts-frontend | 2:30:40 PM [vite] http proxy error at /assistants/public/: opengpts-frontend | Error: connect ECONNREFUSED 172.19.0.4:8000 opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16) opengpts-frontend | 2:30:40 PM [vite] http proxy error at /assistants/: opengpts-frontend | Error: connect ECONNREFUSED 172.19.0.4:8000 opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16) opengpts-frontend | 2:30:40 PM [vite] http proxy error at /runs/config_schema: opengpts-frontend | Error: connect ECONNREFUSED 172.19.0.4:8000 opengpts-frontend | at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1278:16)
@felipeggrod try removing PROXY_URL=your_proxy_url from your .env file. Worked for me.
