When the "docker logs -f ragflow-server" command is run in offline deployment, the request address times out

[liukx362330]

Is there an existing issue for the same bug?

• I have checked the existing issues.

Branch name

main

Commit ID

081f922

Other environment information

No response

Actual behavior

pull down the image from the Internet, and report this connection to a website timeout when migrating to the Intranet for installation

Expected behavior

I'd like to deploy it on the Intranet

Steps to reproduce

load the related image and copy the source code, the first step: cd ragflow/docker, the second step: chmod +x./entrypoint. docker compose- f docker-compose-CN.yml up -d, Step 4: docker logs -f ragflow-server

Additional information

Traceback (most recent call last):
File "/ragflow/api/ragflow_server.py", line 26, in
from api.apps import app
File "/ragflow/api/apps/init.py", line 26, in
from api.db.db_models import close_connection
File "/ragflow/api/db/db_models.py", line 32, in
from api.settings import DATABASE, stat_logger, SECRET_KEY
File "/ragflow/api/settings.py", line 35, in
from rag.utils.es_conn import ELASTICSEARCH
File "/ragflow/rag/utils/init.py", line 59, in
encoder = tiktoken.encoding_for_model("gpt-3.5-turbo")
File "/usr/local/lib/python3.10/dist-packages/tiktoken/model.py", line 101, in encoding_for_model
return get_encoding(encoding_name_for_model(model_name))
File "/usr/local/lib/python3.10/dist-packages/tiktoken/registry.py", line 73, in get_encoding
enc = Encoding(**constructor())
File "/usr/local/lib/python3.10/dist-packages/tiktoken_ext/openai_public.py", line 72, in cl100k_base
mergeable_ranks = load_tiktoken_bpe(
File "/usr/local/lib/python3.10/dist-packages/tiktoken/load.py", line 147, in load_tiktoken_bpe
contents = read_file_cached(tiktoken_bpe_file, expected_hash)
File "/usr/local/lib/python3.10/dist-packages/tiktoken/load.py", line 64, in read_file_cached
contents = read_file(blobpath)
File "/usr/local/lib/python3.10/dist-packages/tiktoken/load.py", line 25, in read_file
resp = requests.get(blobpath)
File "/usr/local/lib/python3.10/dist-packages/requests/api.py", line 73, in get
return request("get", url, params=params, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/requests/api.py", line 59, in request
return session.request(method=method, url=url, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/requests/sessions.py", line 589, in request
resp = self.send(prep, **send_kwargs)
File "/usr/local/lib/python3.10/dist-packages/requests/sessions.py", line 703, in send
r = adapter.send(request, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/requests/adapters.py", line 507, in send
raise ConnectTimeout(e, request=request)
requests.exceptions.ConnectTimeout: HTTPSConnectionPool(host='openaipublic.blob.core.windows.net', port=443):
Max retries exceeded with url: /encodings/cl100k_base.tiktoken (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x7f4dd75afac0>,
'Connection to openaipublic.blob.core.windows.net timed out. (connect timeout=None)'))

[dingsa0210]

Tokenizer model should not be hardcoded in line:
tiktoken.encoding_for_model("gpt-3.5-turbo")

[ye-jeck]

have you solved it? I also encountered this problem

[liwan14x]

may be you can try to add proxy to the docker yml file.
environment:

• TZ=${TIMEZONE}
• HF_ENDPOINT=https://hf-mirror.com/
• HTTP_PROXY=
• HTTPS_PROXY=
• NO_PROXY=