httpx Python 3.7: RuntimeError: read() called while another

The Fix

Upgrade to version 0.13.0 or later.

Based on closed encode/httpx issue #527 · 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

@@ -51,6 +51,7 @@ def __init__(
         self.stream_writer = stream_writer
         self.timeout = timeout
+        self.read_lock = asyncio.Lock()
 
         self._inner: typing.Optional[SocketStream] = None

repro.py

import asyncio
import pickle

import httpx


class SendRequests:
    num_parallel_requests = 0
    MAX_PARALLEL = 100

    async def embedd_batch(self):
        requests = []
        async with httpx.AsyncClient(
                timeout=httpx.TimeoutConfig(timeout=60),
                base_url="https://api.garaza.io/") as client:
            for i in range(1000):
                requests.append(self._send_to_server(client))

            embeddings = await asyncio.gather(*requests)

        return embeddings

    async def __wait_until_released(self):
        while self.num_parallel_requests >= self.MAX_PARALLEL:
            await asyncio.sleep(0.1)

    async def _send_to_server(self, client):
        await self.__wait_until_released()

        self.num_parallel_requests += 1
        # simplified image loading 
        with open("image.pkl", "rb") as f:
            im = pickle.load(f)
        url = "/image/inception-v3?machine=1&session=1&retry=0"
        emb = await self._send_request(client, im, url)
        self.num_parallel_requests -= 1
        return emb

    async def _send_request(self, client, image, url):
        headers = {'Content-Type': 'image/jpeg',
                   'Content-Length': str(len(image))}
        response = await client.post(url, headers=headers, data=image)
        print(response.content)
        return response


if __name__ == "__main__":
    cl = SendRequests()
    asyncio.get_event_loop().run_until_complete(cl.embedd_batch())

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\nUpgrade to version 0.13.0 or later.\nWhen NOT to use: This fix is not suitable for single-threaded applications or non-HTTP/2 contexts.\n\n

Why This Fix Works in Production

Trigger: asyncio.get_event_loop().run_until_complete(cl.embedd_batch())
Mechanism: A race condition caused a RuntimeError when reading streams concurrently in HTTP/2
Why the fix works: Fixes a race condition that caused a RuntimeError when reading streams concurrently in HTTP/2. (first fixed release: 0.13.0).

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

Shows up under Python 3.7 in real deployments (not just unit tests).
A race condition caused a RuntimeError when reading streams concurrently in HTTP/2
Surfaces as: Traceback (most recent call last):

Proof / Evidence

GitHub issue: #527
Fix PR: https://github.com/encode/httpx/pull/535
First fixed release: 0.13.0
Reproduced locally: No (not executed)
Last verified: 2026-02-09
Confidence: 0.75
Did this fix it?: Yes (upstream fix exists)
Own content ratio: 0.22

Discussion

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

Jump to Sources Open on GitHub

“@iluxonchik posted another example in #382, which I'm reposting here in a runnable form: This program runs fine on my side, on 3.7.3 and 3.8.0.…”

@florimondmanca · 2019-11-17 · source

“@PrimozGodec I was able to run your code and reproduce the error you listed (on 3.8.0)”

@florimondmanca · 2019-11-17 · source

“@iluxonchik Thanks for the detailed debugging material”

@florimondmanca · 2019-11-17 · source

“The read() error seems to be solved. I already go more responses when I run code from my example above. With the same script (code…”

@PrimozGodec · 2019-11-18 · source

Failure Signature (Search String)

asyncio.get_event_loop().run_until_complete(cl.embedd_batch())

Error Message

Stack trace

error.txt

Error Message
-------------
Traceback (most recent call last):
  File "/Users/primoz/PycharmProjects/orange3-imageanalytics/example.py", line 63, in <module>
    asyncio.get_event_loop().run_until_complete(cl.embedd_batch())
  File "/Users/primoz/miniconda3/lib/python3.7/asyncio/base_events.py", line 584, in run_until_complete
    return future.result()
  File "/Users/primoz/PycharmProjects/orange3-imageanalytics/example.py", line 18, in embedd_batch
    embeddings = await asyncio.gather(*requests)
  File "/Users/primoz/PycharmProjects/orange3-imageanalytics/example.py", line 41, in _send_to_server
    emb = await self._send_request(client, im, url)
  File "/Users/primoz/PycharmProjects/orange3-imageanalytics/example.py", line 56, in _send_request
    data=image
  File "/Users/primoz/venv/orange/lib/python3.7/site-packages/httpx/client.py", line 484, in post
    trust_env=trust_env,
  File "/Users/primoz/venv/orange/lib/python3.7/site-packages/httpx/client.py", line 626, in request
    trust_env=trust_env,
  File "/Users/primoz/venv/orange/lib/python3.7/site-packages/httpx/client.py", line 650, in send
    trust_env=trust_env,
  File "/Users/primoz/venv/orange/lib/python3.7/site-packages/httpx/client.py", line 265, in _get_response
    return await get_response(request)
  File "/Users/primoz/venv/orange/lib/python3.7/site-packages/httpx/middleware/redirect.py", line 31, in __call__
    response = await ge
... (truncated) ...

Stack trace

error.txt

Error Message
-------------
unhandled exception during asyncio.run() shutdown
task: <Task finished name='Task-115' coro=<print_content_from_url() done, defined at demo.py:4> exception=KeyError(HTTPConnection(origin=Origin(scheme='https' host='www.google.com' port=443)))>
Traceback (most recent call last):
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages/httpx/dispatch/connection_pool.py", line 120, in send
    response = await connection.send(
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages/httpx/dispatch/connection.py", line 62, in send
    response = await self.h2_connection.send(request, timeout=timeout)
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages/httpx/dispatch/http2.py", line 57, in send
    status_code, headers = await self.receive_response(stream_id, timeout)
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages/httpx/dispatch/http2.py", line 176, in receive_response
    event = await self.receive_event(stream_id, timeout)
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages/httpx/dispatch/http2.py", line 211, in receive_event
    data = await self.stream.read(self.READ_NUM_BYTES, timeout, flag=flag)
  File "/Users/iluxonchik/.local/share/virtualenvs/tmp-agwWamBd/lib/python3.8/site-packages
... (truncated) ...

Minimal Reproduction

repro.py

import asyncio
import pickle

import httpx


class SendRequests:
    num_parallel_requests = 0
    MAX_PARALLEL = 100

    async def embedd_batch(self):
        requests = []
        async with httpx.AsyncClient(
                timeout=httpx.TimeoutConfig(timeout=60),
                base_url="https://api.garaza.io/") as client:
            for i in range(1000):
                requests.append(self._send_to_server(client))

            embeddings = await asyncio.gather(*requests)

        return embeddings

    async def __wait_until_released(self):
        while self.num_parallel_requests >= self.MAX_PARALLEL:
            await asyncio.sleep(0.1)

    async def _send_to_server(self, client):
        await self.__wait_until_released()

        self.num_parallel_requests += 1
        # simplified image loading 
        with open("image.pkl", "rb") as f:
            im = pickle.load(f)
        url = "/image/inception-v3?machine=1&session=1&retry=0"
        emb = await self._send_request(client, im, url)
        self.num_parallel_requests -= 1
        return emb

    async def _send_request(self, client, image, url):
        headers = {'Content-Type': 'image/jpeg',
                   'Content-Length': str(len(image))}
        response = await client.post(url, headers=headers, data=image)
        print(response.content)
        return response


if __name__ == "__main__":
    cl = SendRequests()
    asyncio.get_event_loop().run_until_complete(cl.embedd_batch())

Environment

Python: 3.7

What Broke

RuntimeError occurs when sending multiple images to the server concurrently.

Why It Broke

A race condition caused a RuntimeError when reading streams concurrently in HTTP/2

Fix Options (Details)

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

Upgrade to version 0.13.0 or later.

When NOT to use: This fix is not suitable for single-threaded applications or non-HTTP/2 contexts.

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

Fix reference: https://github.com/encode/httpx/pull/535

First fixed release: 0.13.0

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

When NOT to Use This Fix

This fix is not suitable for single-threaded applications or non-HTTP/2 contexts.

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
0.13.0	Fixed

Related Issues

No related fixes found.

Cluster: httpx:race-condition All hubs All clusters

Related clusters: Configuration error Data consistency Other

Sources

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

httpx Python 3.7: RuntimeError: read() called while another coroutine is already waiting for (Fix)

The Fix

Why This Fix Works in Production

Why This Breaks in Prod

Proof / Evidence

Discussion

Failure Signature (Search String)

Error Message

Minimal Reproduction

Environment

What Broke

Why It Broke

Fix Options (Details)

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

Get updates

When NOT to Use This Fix

Verify Fix

Did This Fix Work in Your Case?

Prevention

Version Compatibility Table

Related Issues

Sources