我在Mac上用shadow racket代理可以访问chatgpt，我应该怎么设置docker才能让docker走shadowracket代理访问chatgpt？ #51
Comments
|
|
|
|
|
|
以上是我所有的设置，但是仍显示错误，不知道问题出在哪 |
|
我觉得应该需要修改两个地方：
另外，我有个疑问，为什么这个 docker desktop （好像是吧）显示的代理： 与 docker-compose.yml 里配置设置的不一样： |
|
按照您的意见修改完后仍然报错 |
|
我尝试ping |
|
ping 172.17.0.1:1000 |
nc localhost 1000 -v
# 显示 "Connection to ... succeeded!" 则成功或 telnet localhost 1000
# 显示 "Connected to ..." 则成功
nc host.docker.internal 1000 -v或 telnet host.docker.internal 1000看看能否成功连接。 (万一 nc 和 telnet 都没有的话随便装一个吧) |
|
telnet localhost 1000 显示失败了 |
|
那说明你的代理软件没有配置好吧 |
|
|
在哪里设置呢 |
|
再改回 127.0.0.1 试试？😅 |
|
我吧端口重置 为原来的，就"Connection to ... succeeded!" |
|
那你把 |
|
environment: 这样设置还是报错诶 |
|
During handling of the above exception, another exception occurred: |
然后这个还是不一样
好奇怪
|
|
你重启服务组了没？重新 |
|
重组之后代理显示正常，但还是报错 |
|
你在容器里试试能 telnet 连到代理吗 |
|
(base) zjk-macminim1@Zjk-MacdeMac-mini muvtuber % telnet host.docker.internal 1082 |
|
(base) zjk-macminim1@Zjk-MacdeMac-mini muvtuber % nc host.docker.internal 1082 -v |
啊？这不是容器内叭 🥺 你试试这种炼金去的： docker compose exec chatgpt_chatbot sh |
|
(base) zjk-macminim1@Zjk-MacdeMac-mini muvtuber % docker compose exec chatgpt_chatbot sh telnet host.docker.internal 1082telnet: could not resolve host.docker.internal/1082: Name or service not known 显示错误 |
|
这个 host.docker.internal 按照 docker desktop 的文档应该是可用的。 所以还是代理设置的问题吧，我还是觉得代理要监听 0.0.0.0 应该才能访问到，你再试试改一下？（改完要重启代理软件哦） |
|
(base) zjk-macminim1@Zjk-MacdeMac-mini muvtuber % docker compose exec chatgpt_chatbot sh telnet host.docker.internal 1082telnet: could not resolve host.docker.internal/1082: Name or service not known 还是显示错误诶 |
|
你在宿主机开一个其他服务，在容器内能访问到嘛 比如宿主机uh： 然后在里： |
|
您使用的代理软件是什么，或许我 change 一下软件就行了？ |
b x不行诶
|
|
Docker desktop 文档不是说这个地址是可用的嘛 😭（我自己是用的 colima 所以我不清楚 docker desktop 的具体情况 emmm，开 python 服务之后 mac 弹的那个允许传入网络连接对话框允许了没 |
|
允许了的 |
|
太难了吧，宝子 |
|
这个网络就是很烦，每个人的情况都不一样。解决这个办法挺多的，主要不知道你用啥方案比较方便 你在容器里和 docker 的 vm 里分别 |
|
请问应该怎么查呢 |
|
我ifconfig (base) zjk-macminim1@Zjk-MacdeMac-mini ~ % ifconfig |
|
(base) zjk-macminim1@Zjk-MacdeMac-mini ~ % host host.docker.internal |
|
我自己常用的一种手段是在容器内开一个陷阱服务，然后从宿主机去访问这个服务，便记录下了容器眼里宿主机的 ip。大概就这样吧。 除了找出 ip 去连，这个问题其他解决方案也很多：
还有，想起来以前有个老哥说他是这么解决的：
|
|
i get it ,通过docker desktop的settings-resources-proxies写了宿主机的ip(我写的就是宿主机以太网下的本地ipv4地址),但是得把dockerfile和docker compose 2个文件里的代理都删掉！！！！感谢大佬，晚安 
|
我是把他注释掉了 |
|
太好了🥹 但是为啥，这个具体是怎么个配法我还是没理解（我没用docker desktop不了解）。
通过docker desktop的settings-resources-proxies写了宿主机的ip(我写的就是宿主机以太网下的本地ipv4地址),
这个是怎么配的呢？我想了解一下在文档里详细写一写。
或者如果你能帮忙 pr 改一改 README 就更棒了🥰
|
|
因为我们想让docker使用宿主机的代理，所以我们在docker desktop里settings-resources-proxies设置代理地址（宿主机以太网下的本地ipv4地址+代理软件设置的端口），同时把dockerfile和doccker compose 2个文件中的有关proxy的内容注释掉或删除，这个或许会影响docker 走宿主机的代理，最后感谢大佬的提醒[玫瑰][玫瑰][玫瑰] |
|
哦哦，是 Settings -> Resources -> Proxies 的意思啊。这下看懂了。我之前以为是在某个神秘的地方去写某个 |
|
这样设置对小白很友好🥰 |
2023-06-05 22:01:07
2023-06-05 22:01:07 During handling of the above exception, another exception occurred:
2023-06-05 22:01:07
2023-06-05 22:01:07 Traceback (most recent call last):
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/adapters.py", line 489, in send
2023-06-05 22:01:07 resp = conn.urlopen(
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/urllib3/connectionpool.py", line 787, in urlopen
2023-06-05 22:01:07 retries = retries.increment(
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/urllib3/util/retry.py", line 592, in increment
2023-06-05 22:01:07 raise MaxRetryError(_pool, url, error or ResponseError(cause))
2023-06-05 22:01:07 urllib3.exceptions.MaxRetryError: HTTPSConnectionPool(host='openaipublic.blob.core.windows.net', port=443): Max retries exceeded with url: /encodings/cl100k_base.tiktoken (Caused by ProxyError('Cannot connect to proxy.', NewConnectionError('<urllib3.connection.HTTPSConnection object at 0xffffa80f2230>: Failed to establish a new connection: [Errno -2] Name or service not known')))
2023-06-05 22:01:07
2023-06-05 22:01:07 During handling of the above exception, another exception occurred:
2023-06-05 22:01:07
2023-06-05 22:01:07 Traceback (most recent call last):
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/grpc/_server.py", line 444, in _call_behavior
2023-06-05 22:01:07 response_or_iterator = behavior(argument, context)
2023-06-05 22:01:07 File "/app/chatgpt/grpcapi.py", line 37, in NewSession
2023-06-05 22:01:07 session_id = self.multiChatGPT.new_session(config)
2023-06-05 22:01:07 File "/app/chatgpt/chatbot.py", line 289, in new_session
2023-06-05 22:01:07 self.chatgpts[session_id] = ChatGPTProxy(
2023-06-05 22:01:07 File "/app/chatgpt/chatbot.py", line 189, in init
2023-06-05 22:01:07 self.renew()
2023-06-05 22:01:07 File "/app/chatgpt/chatbot.py", line 193, in renew
2023-06-05 22:01:07 self.chatgpt = self._new_chatgpt(self.config)
2023-06-05 22:01:07 File "/app/chatgpt/chatbot.py", line 207, in _new_chatgpt
2023-06-05 22:01:07 new_chatgpt = ChatGPTv3(config={
2023-06-05 22:01:07 File "/app/chatgpt/chatbot.py", line 94, in init
2023-06-05 22:01:07 self.chatbot = ChatbotV3(
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/revChatGPT/V3.py", line 68, in init
2023-06-05 22:01:07 if self.get_token_count("default") > self.max_tokens:
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/revChatGPT/V3.py", line 116, in get_token_count
2023-06-05 22:01:07 encoding = tiktoken.encoding_for_model(self.engine)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken/model.py", line 75, in encoding_for_model
2023-06-05 22:01:07 return get_encoding(encoding_name)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken/registry.py", line 63, in get_encoding
2023-06-05 22:01:07 enc = Encoding(**constructor())
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken_ext/openai_public.py", line 64, in cl100k_base
2023-06-05 22:01:07 mergeable_ranks = load_tiktoken_bpe(
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken/load.py", line 114, in load_tiktoken_bpe
2023-06-05 22:01:07 contents = read_file_cached(tiktoken_bpe_file)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken/load.py", line 46, in read_file_cached
2023-06-05 22:01:07 contents = read_file(blobpath)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/tiktoken/load.py", line 24, in read_file
2023-06-05 22:01:07 return requests.get(blobpath).content
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/api.py", line 73, in get
2023-06-05 22:01:07 return request("get", url, params=params, **kwargs)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/api.py", line 59, in request
2023-06-05 22:01:07 return session.request(method=method, url=url, **kwargs)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/sessions.py", line 587, in request
2023-06-05 22:01:07 resp = self.send(prep, **send_kwargs)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/sessions.py", line 701, in send
2023-06-05 22:01:07 r = adapter.send(request, **kwargs)
2023-06-05 22:01:07 File "/usr/local/lib/python3.10/site-packages/requests/adapters.py", line 559, in send
2023-06-05 22:01:07 raise ProxyError(e, request=request)
2023-06-05 22:01:07 requests.exceptions.ProxyError: HTTPSConnectionPool(host='openaipublic.blob.core.windows.net', port=443): Max retries exceeded with url: /encodings/cl100k_base.tiktoken (Caused by ProxyError('Cannot connect to proxy.', NewConnectionError('<urllib3.connection.HTTPSConnection object at 0xffffa80f2230>: Failed to establish a new connection: [Errno -2] Name or service not known')))
The text was updated successfully, but these errors were encountered: