Pydantic Dataclass instances are missing attributes at

The Fix

pip install pydantic==2.12.5

Based on closed pydantic/pydantic issue #12505 · PR/commit linked

Production note: This tends to surface only under concurrency. Reproduce with load tests and watch for lock contention/cancellation paths.

Jump to Verify Open PR/Commit

@@ -11,7 +11,7 @@
 from typing_extensions import TypeGuard, dataclass_transform
 
-from ._internal import _config, _decorators, _namespace_utils, _typing_extra
+from ._internal import _config, _decorators, _mock_val_ser, _namespace_utils, _typing_extra
 from ._internal import _dataclasses as _pydantic_dataclasses

repro.py

from __future__ import annotations
import threading

from pandera import DataFrameModel
from pandera.typing import DataFrame
from pydantic.dataclasses import dataclass


@dataclass
class Base:
    """
    Dropping this inheritance will result in the following error:

      Thread 1: 'MyObject' object has no attribute '__pydantic_validator__'

    occuring during object instantiation instead of

      Thread 1: 'MyObject' object has no attribute 'my_attr_1'

    during attribute access.
    """

    pass


@dataclass
class MyObject(Base):
    my_attr_1: MyNestedObject | None
    my_attr_2: DataFrame[DataFrameModel] | None


@dataclass
class MyNestedObject:
    pass


def create_obj(thread_id: int, results: list, errors: list) -> None:
    try:
        obj = MyObject(my_attr_1=None, my_attr_2=None)
        _ = obj.my_attr_1
    except Exception as e:
        errors.append((thread_id, e))
    else:
        results.append(thread_id)


def main():
    results = []
    errors = []

    threads = []
    for i in range(10):
        t = threading.Thread(target=create_obj, args=(i, results, errors))
        threads.append(t)

    for t in threads:
        t.start()

    for t in threads:
        t.join()

    print(f"Results: {len(results)} successful, {len(errors)} errors")
    print()

    if errors:
        print("Errors encountered:")
        for thread_id, error in errors:
            print(f"  Thread {thread_id}: {error}")
        print()


if __name__ == "__main__":
    exit(main())

verify

Re-run the minimal reproduction on your broken version, then apply the fix and re-run.

fix.md

Option A — Upgrade to fixed release\npip install pydantic==2.12.5\nWhen NOT to use: This fix should not be used if the application does not involve concurrent requests.\n\n

Why This Fix Works in Production

Trigger: Dataclass instances are missing attributes at runtime
Mechanism: Race condition during concurrent requests leads to incomplete initialization of dataclass instances
Why the fix works: Prevents the deletion of the mock validator/serializer in `rebuild_dataclass()`, addressing a race condition that caused missing attributes in dataclass instances. (first fixed release: 2.12.5).

Production impact:

If left unfixed, failures can be intermittent under concurrency (hard to reproduce; shows up as sporadic 5xx/timeouts).

Why This Breaks in Prod

Race condition during concurrent requests leads to incomplete initialization of dataclass instances
Production symptom (often without a traceback): Dataclass instances are missing attributes at runtime

Proof / Evidence

GitHub issue: #12505
Fix PR: https://github.com/pydantic/pydantic/pull/12513
First fixed release: 2.12.5
Reproduced locally: No (not executed)
Last verified: 2026-02-09
Confidence: 0.95
Did this fix it?: Yes (upstream fix exists)
Own content ratio: 0.42

Discussion

High-signal excerpts from the issue thread (symptoms, repros, edge-cases).

Jump to Sources Open on GitHub

“### Initial Checks - [x] I confirm that I'm using Pydantic V2 ### Description Hi everyone, I've encountered a strange bug that was quite hard to track down. The symptom was that accessing a certain attribute of an object (an instance of pyd”

Issue thread · issue description · source

Failure Signature (Search String)

Dataclass instances are missing attributes at runtime
I still don't completely understand the details. It seems to me that it's a race condition that leads to incomplete initialization, but I haven't been able to make more headway

Copy-friendly signature

signature.txt

Failure Signature
-----------------
Dataclass instances are missing attributes at runtime
I still don't completely understand the details. It seems to me that it's a race condition that leads to incomplete initialization, but I haven't been able to make more headway into the problem.

Error Message

Signature-only (no traceback captured)

error.txt

Error Message
-------------
Dataclass instances are missing attributes at runtime
I still don't completely understand the details. It seems to me that it's a race condition that leads to incomplete initialization, but I haven't been able to make more headway into the problem.

Minimal Reproduction

repro.py

from __future__ import annotations
import threading

from pandera import DataFrameModel
from pandera.typing import DataFrame
from pydantic.dataclasses import dataclass


@dataclass
class Base:
    """
    Dropping this inheritance will result in the following error:

      Thread 1: 'MyObject' object has no attribute '__pydantic_validator__'

    occuring during object instantiation instead of

      Thread 1: 'MyObject' object has no attribute 'my_attr_1'

    during attribute access.
    """

    pass


@dataclass
class MyObject(Base):
    my_attr_1: MyNestedObject | None
    my_attr_2: DataFrame[DataFrameModel] | None


@dataclass
class MyNestedObject:
    pass


def create_obj(thread_id: int, results: list, errors: list) -> None:
    try:
        obj = MyObject(my_attr_1=None, my_attr_2=None)
        _ = obj.my_attr_1
    except Exception as e:
        errors.append((thread_id, e))
    else:
        results.append(thread_id)


def main():
    results = []
    errors = []

    threads = []
    for i in range(10):
        t = threading.Thread(target=create_obj, args=(i, results, errors))
        threads.append(t)

    for t in threads:
        t.start()

    for t in threads:
        t.join()

    print(f"Results: {len(results)} successful, {len(errors)} errors")
    print()

    if errors:
        print("Errors encountered:")
        for thread_id, error in errors:
            print(f"  Thread {thread_id}: {error}")
        print()


if __name__ == "__main__":
    exit(main())

Environment

Pydantic: 2

What Broke

Accessing attributes of dataclass instances results in exceptions due to missing attributes.

Why It Broke

Race condition during concurrent requests leads to incomplete initialization of dataclass instances

Fix Options (Details)

Option A — Upgrade to fixed release Safe default (recommended)

pip install pydantic==2.12.5

When NOT to use: This fix should not be used if the application does not involve concurrent requests.

Use when you can deploy the upstream fix. It is usually lower-risk than long-lived workarounds.

Fix reference: https://github.com/pydantic/pydantic/pull/12513

First fixed release: 2.12.5

Last verified: 2026-02-09. Validate in your environment.

When NOT to Use This Fix

This fix should not be used if the application does not involve concurrent requests.

Verify Fix

verify

Re-run the minimal reproduction on your broken version, then apply the fix and re-run.

Did This Fix Work in Your Case?

Quick signal helps us prioritize which fixes to verify and improve.

Prevention

Add a CI check that diffs key outputs after upgrades (OpenAPI schema snapshots, JSON payload shapes, CLI output).
Upgrade behind a canary and run integration tests against the canary before 100% rollout.
Add a stress test that runs high-concurrency workloads and fails on thread dumps / blocked locks.
Enable watchdog dumps in prod (faulthandler, thread dump endpoint) to capture deadlocks quickly.

Version Compatibility Table

Version	Status
2.12.5	Fixed

Related Issues

No related fixes found.

Cluster: pydantic:race-condition Pydantic hub Pydantic best practices All hubs All clusters

Related clusters: Configuration error Data consistency OpenAPI schema

Sources

We don’t republish the full GitHub discussion text. Use the links above for context.

Pydantic Dataclass instances are missing attributes at runtime (Fix)