3 months ago
Description: Open-source AI platform with BaaS & LLMOps for all users.
Category: AI/ML
URL: https://railway.com/template/TiT2V6
10 Replies
3 months ago
You will find the initiate password in the Api service in the INIT_PASSWORD environment variable
3 months ago
Just updated the template to support version v1.0.1
a month ago
Hello! Thank you very much for your contribution! I believe that using dify.ai by railway is very practical. I came across an error that I noticed in other templates. The knowledge base is unable to complete a successful embedding. Do you know how to resolve this? Thank you very much!
cledsonsantana
Hello! Thank you very much for your contribution! I believe that using dify.ai by railway is very practical. I came across an error that I noticed in other templates. The knowledge base is unable to complete a successful embedding. Do you know how to resolve this? Thank you very much!
a month ago
Hi @cledsonsantana do you have this issue also in this template? can you share some screenshots and a Dify workflow export so I can reproduce it and look into it? thanks
a month ago
Hey @Ramarail, I also have this error, everytime I try embedding it I see this error in railway:
File "/app/api/core/provider_manager.py", line 123, in get_configurations
raise MaxRetryError(_pool, url, reason) from reason # type: ignore[arg-type]
provider_entities = model_provider_factory.get_providers()
File "/app/api/core/plugin/impl/base.py", line 69, in _request
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
raise PluginDaemonInnerError(code=-500, message="Request to Plugin Daemon Service failed")
File "/app/api/core/model_runtime/model_providers/model_provider_factory.py", line 59, in get_providers
urllib3.exceptions.MaxRetryError: HTTPConnectionPool(host='localhost', port=5002): Max retries exceeded with url: /plugin/e1c150e7-a130-4701-b162-7b67ad6ff345/management/models?page=1&page_size=256 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f0a24606d80>: Failed to establish a new connection: [Errno 111] Connection refused'))
core.plugin.entities.plugin_daemon.PluginDaemonInnerError
plugin_providers = self.get_plugin_model_providers()
2025-05-23 13:43:51,721.721 INFO [Dummy-4] [retry_document_indexing_task.py:102] - Retry dataset: 1a620e09-c397-451e-90bf-0f8953df4f39 latency: 0.8641215790994465
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
During handling of the above exception, another exception occurred:
File "/app/api/core/model_runtime/model_providers/model_provider_factory.py", line 95, in get_plugin_model_providers
Traceback (most recent call last):
plugin_providers = self.plugin_model_manager.fetch_model_providers(self.tenant_id)
File "/app/api/core/plugin/impl/base.py", line 64, in _request
response = requests.request(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/api/core/plugin/impl/model.py", line 29, in fetch_model_providers
response = self._request_with_plugin_daemon_response(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/api/core/plugin/impl/base.py", line 139, in requestwith_plugin_daemon_response
response = self._request(method, path, headers, data, params, files)
ramarail
Hi @cledsonsantana do you have this issue also in this template? can you share some screenshots and a Dify workflow export so I can reproduce it and look into it? thanks
a month ago
I posted the trace above, not sure if you saw the other message. Thanks for the template!
a month ago
Hi Matt in which service are you getting this message? and can you share the steps to reproduce so I can look into it?
a month ago
Sure, I just deployed your template, then I created a knowledge base in the UI and tried adding a website with Jina. The error occurs in the worker. I have also added openai.
Attachments
matthiasriener
Sure, I just deployed your template, then I created a knowledge base in the UI and tried adding a website with Jina. The error occurs in the worker. I have also added openai.
a month ago
Thanks for the clarification. I have found the issue and fixed it in the template and for the current environment you are running just copy paste the following variables value as is in your worker variables and redeploy the worker and that should solve the issue. Ping me if it doesn't.
PLUGIN_DAEMON_KEY="${{plugin-daemon.SERVER_KEY}}"
PLUGIN_DAEMON_URL="http://${{plugin-daemon.RAILWAY_PRIVATE_DOMAIN}}:5002"
a month ago
That seemed to have fixed the issue. Thanks a lot for the help and the template!