Disable third party telemetry

[Yanni8]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

Some third-party integrations (for example, ChromaDB) collect telemetry data.

It is always annoying to search for all third-party libraries that langchain uses and see if they collect any data or not.

It would be great to have an environment variable that could disable all telemetry data at once (probably not possible but would be nice). Or alternatively, there a page in the documentation where all third-party integrations are listed that collect telemetry data with an tutorial on how to disable data collection.

Motivation

It would be nice to have the possibility of easily finding out which library collects data. This would have two benefits. It would save the developer a lot of time, and it would minimize the risk of missing a library and accidentally sending data to them.

It would even be better if there would be a simple functionality that gives the user the possibility to easily disable all data collection in all thirt party libraries.

Proposal (If applicable)

No response

[eyurtsev]

What telemetry data are you concerned about specifically? Are you seeing any information besides information about langchain version etc? Could you include a few references of things you find concerning?

[fremontfancy]

@eyurtsev
If we want to deploy the LLM + chromeDB as private (without connecting to internet) chatbot. This telemetry call is annoying and delaying each and every calls to database. It tries to reach posthog.com api and holds the response until it times out. Everywhere it is mentioned.. how to turn off (this telemetry call) using chromadb pkg, but unable to find any solutions for langchain (Chroma) version.

INFO:backoff:Backing off send_request(...) for 0.2s (requests.exceptions.ConnectTimeout: HTTPSConnectionPool(host='us-api.i.posthog.com', port=443): Max retries exceeded with url: [/batch/](https://datalab-aws.corp.apple.com/batch/) (Caused by ConnectTimeoutError(<urllib3.connection.HTTPSConnection object at 0x7f529ca1c050>, 'Connection to us-api.i.posthog.com timed out. (connect timeout=15)')))

[fremontfancy]

@Yanni8 This solution works for me, I am not seeing any more Telemetry calls.

....
....

loader = TextLoader("my_data.txt")
docs = loader.load()
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
documents = text_splitter.split_documents(docs)

embeddings = OllamaEmbeddings(model="mistral")
client_settings = chromadb.config.Settings(
    is_persistent=True,
    persist_directory='my_dir',
    anonymized_telemetry=False,
)
vector=Chroma(
    collection_name = 'my_collection',
    client_settings=client_settings,
    embedding_function=embeddings,
)

vector.add_documents(documents)

....
....