Phoenix and multiple webhooks ignored

xgeek116

Hello,

I have an elixir phoneix application that receives Stripe webhooks to update payment status in the database, I have a problem, sometimes my webhook endpoint does not return a response to stripe so the webhook will be failed.

When I looked in the logs : I found that sometimes the request does not pass through the Plug (in the Plug I check the signature of the webhook), and then the request can reach the Controller function.

The Plug code :


def call(%{request_path: "/webhooks/stripe"} = conn, _) do
    check_signature(conn, "secret")
  end


  defp check_signature(conn, signing_secret) do
      # Stripe.Webhook.construct_event(body, stripe_signature, signing_secret)
      # etc..
      IO.puts("signature OK")
      Plug.Conn.assign(conn, :stripe_event, stripe_event)
    else
      {:error, error} ->
        Logger.info(error)
        conn
        |> send_resp(:bad_request, "Stripe signature error: #{error}")
        |> halt()
      _ -> conn
      |> send_resp(:bad_request, "Stripe signature error")
      |> halt()
    end
  end

So in the logs the request can be terminated in the Plug and the last log entry is “signature OK”

Why do you think is that ?

8 comments

/phoenix #webhooks #troubleshooting

1 1007 8

2025-05-25 11:38:06 UTC

First 8 of 8 Posts

dimitarvp

Well it seems like the plug is executed successfully then. You’ll have to post more code that is supposed to be executed after it to troubleshoot further.

Post #1

xgeek116

Here is the function in the controller :

  def webhooks_payment(%Plug.Conn{assigns: %{stripe_event: stripe_event}} = conn, _params) do
    IO.puts("webhooks_payment controller reached")
    case handle_payment_intent_event(stripe_event) do
      {:ok, _result} ->
        IO.puts("webhooks_payment controller ok")
        handle_success(conn)

      {:error, error} ->
        IO.puts("webhooks_payment controller error")
        Logger.error(error)
        handle_error(conn, error)
    end
  end

I put 2 screenshots to show how sometimes it can reach the controller function and sometimes it doesn’t even log the entry message.

If the app is working on the request (still in Plug) and another webhook is received, will that block the current one ?

Post #2

xgeek116

What do you think @dimitarvp ? I don’t think it’s a networking problem (k8s istio ingress) becuase the request can reach the Plug but not pass to the Controller

Post #3

xgeek116

Any ideas please ?

Post #4

codeanpeace

That does seem quite mysterious…

Maybe add an IO.inspect(stripe_event) after this or wrap the assignment like so

conn
|> IO.inspect(label: "conn before when signature OK")
|> Plug.Conn.assign(:stripe_event, stripe_event)
|> IO.inspect(label: "conn after when signature OK")

to determine if there’s anything unexpected going on with the conn that do not reach the function in the controller.

The request could also potentially be reaching the controller and just not matching on webhooks_payment if there’s no catchall function head e.g.
def webhooks_payment(conn, params), do: IO.inspect(conn, label: "conn from catchall webhooks_payments")

Post #5

adao

If you followed Stripe Webhooks in Phoenix with Elixir Pattern Matching | Connor Fritz (which it kind of looks like from your function signatures) , I might know the fix . I ran into the same problem when referencing said article .

In Connor’s code , I believe the offending lines are :

{:ok, body, _} = Plug.Conn.read_body(conn)
conn

The conn returned is ‘stale’ in a sense . So I changed this to

{:ok, body, conn} = Plug.Conn.read_body(conn)
conn

and the problem seems to have gone away . I haven’t done any further digging into exactly why this is the case (the exact difference between the conn passed into Plug.Conn.read_body and the one that comes out , and how this affects the pipeline downstream . If anyone has any insight on it , am curious to hear)

Post #6

dimitarvp

While I can’t comment on the exact reasons I’d always advise to use the Plug.Conn that’s returned to you. If you have such a return value then the underlying library has changed it and gives you the modified copy. So just use that.

Post #7

tensor

I’ve had the same problem, luckily enough I have found your comment. Thank you very much.

Digging into the read_body method in Plug codebase, it’s explicitly saying “Like all functions in this module, the conn returned by read_body must be passed to the next stage of your pipeline and should not be ignored.” as it seems the conn returned is being updated while reading the body.

Here is the method implementation and relevant docs plug/lib/plug/conn.ex at 1947edc8171f923ecc262efba6a0946264fa5582 · elixir-plug/plug · GitHub

Post #8

— All posts loaded —