I’m trying to script setup and configuration of caddy server based on a custom download that includes additional plugins (caddy-auth-portal, caddy-auth-jwt, caddy-trace, and various caddy-dns modules ).

During setup, the caddy unit file is configured to run caddy as a non priveledged user (by design).

To get certificates configured properly we are attempting to use the caddy trust command

OS/Environment

None yet, but I'm working on OS X v10.11.6 'El Capitan' client-side. (See 'The way of deployment (cloud or local)' below for details on considered server infrastructure.)

Ansible version

None yet, as I haven't yet gotten that far into setting up Algo.

Version of components from requirements.txt

Not applicable (see above.)

Summary of the pro

On Ubuntu 20.10 x64

sys/sanitize.sh
export ASAN_OPTIONS=detect_odr_violation=0
export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libasan.so.6

Then run tests. Memory leak everywhere.

Describe the bug

Thank you for a great project and documentation. I ran into a bug (I think) while playing with the 5 minute tutorial.

When revoking an access token that is not found the hydra server returns a 404 response but this is not a defined response in the swagger specification. This leads to an error like `response status code does not match any response statuses defined for thi

🎉 Description

The ECDSA contract has a function toEthSignedMessage(bytes32), but we should have a function that works for any length of a bytes array:

function toEthSignedMessage(bytes s) pure internal returns (bytes32) {
  return keccak256(
    "\x19Ethereum Signed Message:\n",
    uintToBytes(s.length),
    s);
}

Where uintToBytes is implemented from an ideally