langchain-ai/langchain · error · RuntimeError

Tracing using LangChainTracerV1 is no longer supported. Plea

Error message

Tracing using LangChainTracerV1 is no longer supported. Please set the LANGCHAIN_TRACING_V2 environment variable to enable tracing instead.

What it means

In `_configure` (manager.py ~2506), if v1 tracing is enabled (`LANGCHAIN_TRACING_V1`/LangChainTracerV1) but v2 is not (`LANGCHAIN_TRACING_V2` unset/false), a `RuntimeError` is raised: LangChainTracerV1 tracing was removed. The library refuses to run with a v1-only configuration rather than silently not tracing.

Source

Thrown at libs/core/langchain_core/callbacks/manager.py:2506

        callback_manager.add_metadata(local_metadata or {}, inherit=False)
    if tracing_tags:
        callback_manager.add_tags(tracing_tags.copy())

    v1_tracing_enabled_ = env_var_is_set("LANGCHAIN_TRACING") or env_var_is_set(
        "LANGCHAIN_HANDLER"
    )

    tracer_v2 = tracing_v2_callback_var.get()
    tracing_v2_enabled_ = _tracing_v2_is_enabled()

    if v1_tracing_enabled_ and not tracing_v2_enabled_:
        # if both are enabled, can silently ignore the v1 tracer
        msg = (
            "Tracing using LangChainTracerV1 is no longer supported. "
            "Please set the LANGCHAIN_TRACING_V2 environment variable to enable "
            "tracing instead."
        )
        raise RuntimeError(msg)

    tracer_project = _get_tracer_project()
    debug = _get_debug()
    if verbose or debug or tracing_v2_enabled_:
        if verbose and not any(
            isinstance(handler, StdOutCallbackHandler)
            for handler in callback_manager.handlers
        ):
            if debug:
                pass
            else:
                callback_manager.add_handler(StdOutCallbackHandler(), inherit=False)
        if debug and not any(
            isinstance(handler, ConsoleCallbackHandler)
            for handler in callback_manager.handlers
        ):
            callback_manager.add_handler(ConsoleCallbackHandler())
        if tracing_v2_enabled_ and not any(

View on GitHub (pinned to e32fa9a52e)

Solutions

1. Set `LANGCHAIN_TRACING_V2=true` (and `LANGCHAIN_API_KEY`, optionally `LANGCHAIN_PROJECT`) — the error is suppressed once v2 is on
2. Remove `LANGCHAIN_TRACING_V1` and any LangChainTracerV1 references entirely
3. Search the repo/deploy configs for the v1 variable: `grep -r LANGCHAIN_TRACING_V1 .`
4. After enabling v2, confirm traces appear in the LangSmith project named by LANGCHAIN_PROJECT

Example fix

# before
LANGCHAIN_TRACING_V1=true
LANGCHAIN_API_KEY=ls__...

# after
LANGCHAIN_TRACING_V2=true
LANGCHAIN_API_KEY=ls__...
LANGCHAIN_PROJECT=my-project

import os

def tracing_config_ok() -> bool:
    v1 = os.environ.get('LANGCHAIN_TRACING_V1')
    v2 = os.environ.get('LANGCHAIN_TRACING_V2')
    return not (v1 and not _truthy(v2))

def _truthy(v):
    return str(v).lower() in {'1', 'true', 'yes'}

When it happens

Trigger: Setting `LANGCHAIN_TRACING_V1=true` (or passing a LangChainTracerV1 via env/callbacks) without `LANGCHAIN_TRACING_V2=true`; legacy deployments with old env var sets; .env files carried over from pre-0.1 projects.

Common situations: Upgrading old projects that used LANGCHAIN_TRACING_V1; CI pipelines with stale env matrices; copied docker-compose/k8s manifests still exporting v1 vars. When both are set, v1 is silently ignored, so the error only appears for v1-only configs.

Related errors

• Invalid format: {self._schema_format}
• Received both `client` and `client_kwargs`. Pass `client_kwa
• Could not resolve content_key {full_path!r}: expected a mapp
• Could not resolve content_key {full_path!r}: missing key {ke
• Trying to deserialize something that cannot be deserialized

AI-assisted analysis of langchain-ai/langchain@e32fa9a52e (2026-08-14). Data as JSON: /api/errors/f5265f0debc2a120. Report an issue: GitHub.