Loading VertexAIEmbeddings is stuck

Gemini API
1

This is my code;

from langchain_google_vertexai import VertexAIEmbeddings
self._embeddings = VertexAIEmbeddings(
    model_name="text-embedding-004"
)

It works well in my local dev env(Docker). But in my production env(GKE, Kubernetes), it stuck on below line intermittently ;

self._embeddings = VertexAIEmbeddings(
    model_name="text-embedding-004"
)

As it works from time to time, I don’t think it’s an auth issue. Does anyone got a similar issue?

2

Hi @JakeJ ,
Can you check if there’s any pod-level logs or micro‑service timeouts around the instantiation or try increasing open file limits or adjusting network retry configs to see if that eases the intermittent blocking?

Thanks!