GoogleGenerativeAIError: Error embedding content: Timeout of 60.0s exceeded, last exception: 503 failed to connect to all addresses; last error: UNKNOWN: ipv4:142.251.43.10:443: socket is null

[3258089958]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain_google_genai import GoogleGenerativeAIEmbeddings

embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
vector = embeddings.embed_query("hello, world!")
vector[:5]

Description

This error occurs:GoogleGenerativeAIError: Error embedding content: Timeout of 60.0s exceeded, last exception: 503 failed to connect to all addresses; last error: UNKNOWN: ipv4:142.251.43.10:443: socket is null

This happened before when I ran the code llm = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.9), but it was fixed when I added transport="rest".
That is llm = ChatGoogleGenerativeAI(model="gemini-pro", transport="rest",temperature=0.9)
This code is working correctly.
I'm trying to figure out what to do with the
"from langchain_google_genai import GoogleGenerativeAIEmbeddings

embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
vector = embeddings.embed_query("hello, world!")
vector[:5]" runs successfully in code like this.

System Info

This error occurs:GoogleGenerativeAIError: Error embedding content: Timeout of 60.0s exceeded, last exception: 503 failed to connect to all addresses; last error: UNKNOWN: ipv4:142.251.43.10:443: socket is null

This happened before when I ran the code llm = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.9), but it was fixed when I added transport="rest".
That is llm = ChatGoogleGenerativeAI(model="gemini-pro", transport="rest",temperature=0.9)
This code is working correctly.
I'm trying to figure out what to do with the
"from langchain_google_genai import GoogleGenerativeAIEmbeddings

embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
vector = embeddings.embed_query("hello, world!")
vector[:5]" runs successfully in code like this.

[dosubot[bot]]

Hey there, @3258089958! Good to see you back in the world of code conundrums. How’s the coding journey treating you this time?

Based on the information you've provided, it seems like you're encountering a GoogleGenerativeAIError when trying to use the GoogleGenerativeAIEmbeddings class in the LangChain framework. You've mentioned that a similar issue was resolved by adding the transport="rest" parameter when using the ChatGoogleGenerativeAI class.

While the context provided does not explicitly show how to pass additional parameters like transport to the GoogleGenerativeAIEmbeddings class, it's possible that the class supports a similar configuration approach as ChatGoogleGenerativeAI. If that's the case, you could try modifying your code as follows:

from langchain_google_genai import GoogleGenerativeAIEmbeddings

# Assuming the class supports a 'transport' parameter similar to ChatGoogleGenerativeAI
embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", transport="rest")
vector = embeddings.embed_query("hello, world!")
vector[:5]

This modification assumes that the GoogleGenerativeAIEmbeddings class's constructor can accept a transport parameter similarly to how it's described for ChatGoogleGenerativeAI in the context. If the class does not support this parameter or requires a different method for specifying the transport method, this approach may not work.

Unfortunately, I wasn't able to find specific information within the LangChain repository about whether the GoogleGenerativeAIEmbeddings class supports a transport parameter, or about alternative methods for handling connection issues. It's also unclear why the class is encountering a timeout error when trying to connect to the server.

If the modification suggested above doesn't resolve the issue, it would be helpful to have more information about the specific circumstances under which the error occurs. For example, does the error occur consistently, or only under certain conditions? Are there any other error messages or logs that might provide more insight into the issue? This additional information could help in identifying the root cause of the problem and suggesting a more targeted solution.

Sources

• docs/docs/integrations/text_embedding/google_generative_ai.ipynb

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[qingzhao0512]

Hi. I came across the same problem. Have you solve this problem? Many thanks~

[XJTUGary]

me, too

[yoursbest]

I encountered the same issue. By adding transport="rest" when constructing the GoogleGenerativeAIEmbeddings didn't resolve the issue. However, transport="rest" works for GoogleGenerativeAI and ChatGoogleGenerativeAI.

[SohamTolwala]

any solutions to this? I am getting the same error but while embedding my text with an embedding model

here is an overview of my problem and what all I have tried:

GoogleGenerativeAIError: Error embedding content: 504 Deadline Exceeded
my code snippet:

def create_subject_vector_db(sub_name):

chunks_docs = load_and_chunk(f"/content/{sub_name}")

CHROMA_DB_DIRECTORY = f'/content/CHROMA_DB_DIRECTORY_{sub_name}'



embedding = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=google_api_key, request_options={'retry':retry.Retry()})

vector_db = Chroma.from_documents(documents=chunks_docs, embedding=embedding, persist_directory=CHROMA_DB_DIRECTORY)

vector_db.persist()

retriever = vector_db.as_retriever()



return retriever

I have tried splitting the load i.e. chunks into more smaller batches, tried the retry method with Exponential Backoff, also tried it running from another deivce (assuming it is a network error)