[BUG] in agent flow, workers can't find an uploaded file

[bodzebod]

Describe the bug
In an agent flow, I have created a pgvector vector store with file upload enabled.
The full file upload is disabled.
I upload a text log file in the chat and ask for analyzing its content, but the workers are unable to find it.
I can see the file was saved under ~/.flowise/storage/78dd4569-306d-4de7-abbb-a90f0ceacf10 .

To Reproduce
Steps to reproduce the behavior:

1. Create an agent flow with a postgres pgvector vector store, a supervisor and a worker
2. in the chat, upload a text file
3. Ask a question about it
4. the supervisor transfers the question to the worker. The worker replies that he doesn't have the file.
   Then the supervisor asks again the same question. Then the worker replies the same again. And so on...

Expected behavior
The workers should be able to find the uploaded file via the chat.

Screenshots

Flow
Test Agent Agents (2).json

Setup

• Installation: docker compose
• Flowise Version: 2.2.3
• OS: Amazon Linux
• Browser: Edge

Additional context
In the screenshots, you see the pgvector is set for a table name "logfiles". After the upload, this table remains desperately empty.
Here is what appears in the debug traces when I upload a file and asks for a question:

2024-12-24 16:00:55 [INFO]: ⬆️ POST /api/v1/vector/internal-upsert/78dd4569-306d-4de7-abbb-a90f0ceacf10
2024-12-24 16:00:55 [DEBUG]: [server]: Initializing Recursive Character Text Splitter (recursiveCharacterTextSplitter_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Finished initializing Recursive Character Text Splitter (recursiveCharacterTextSplitter_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Initializing Ollama Embeddings (ollamaEmbedding_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Finished initializing Ollama Embeddings (ollamaEmbedding_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Initializing Text File (textFile_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Finished initializing Text File (textFile_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Upserting Postgres (postgres_0)
2024-12-24 16:00:55 [DEBUG]: [server]: Finished upserting Postgres (postgres_0)
2024-12-24 16:00:55 [INFO]: [server]: Chatflow 78dd4569-306d-4de7-abbb-a90f0ceacf10 added into ChatflowPool
2024-12-24 16:00:58 [INFO]: ⬆️ POST /api/v1/internal-prediction/78dd4569-306d-4de7-abbb-a90f0ceacf10
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing InMemory Cache (inMemoryCache_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing InMemory Cache (inMemoryCache_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Azure ChatOpenAI (azureChatOpenAI_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Azure ChatOpenAI (azureChatOpenAI_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Postgres Agent Memory (postgresAgentMemory_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Postgres Agent Memory (postgresAgentMemory_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Document Store (Vector) (documentStoreVS_0)
Api key is used with unsecure connection.
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Document Store (Vector) (documentStoreVS_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Supervisor (supervisor_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Supervisor (supervisor_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Retriever Tool (retrieverTool_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Retriever Tool (retrieverTool_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Initializing Worker (worker_0)
2024-12-24 16:00:58 [DEBUG]: [server]: Finished initializing Worker (worker_0)
2024-12-24 16:00:58 [INFO]: [server]: Chatflow 78dd4569-306d-4de7-abbb-a90f0ceacf10_bcec0c53-6c92-44c4-8854-0e0fcf796f08 added into ChatflowPool
2024-12-24 16:00:58 [VERBOSE]: [chain/start] [1:chain:LangGraph] Entering Chain run with input: {
"input": ""
}
2024-12-24 16:00:58 [VERBOSE]: [chain/start] [1:chain:LangGraph > 2:chain:start] Entering Chain run with input: {
"messages": [
{
"lc": 1,
"type": "constructor",
"id": [
"langchain_core",
"messages",
"HumanMessage"
],
"kwargs": {
"content": "what's in this file?",
"additional_kwargs": {},
"response_metadata": {}
}
}
]
}