Rust queue consumer worker gets `memory access out of bounds` exceptions after a couple of hours

A Rust queue consumer worker starts throwing ‘memory access out of bounds’ exceptions for all invocations after a large number of processed events. Changing concurrency doesn’t help, it seems the worker runtime has a memory leak as all I do in the init is set a log shipper.

{
  "outcome": "exception",
  "scriptName": "***",
  "diagnosticsChannelEvents": [],
  "exceptions": [
    {
      "name": "CompileError",
      "message": "memory access out of bounds",
      "timestamp": 1691736709675
    }
  ],
  "logs": [],
  "eventTimestamp": 1691736709675,
  "event": {
    "batchSize": 1,
    "queue": "location-trigger-aggregate"
  },
  "id": 24
}

All I do in the lib.rs file for init is

#[event(start)]
pub fn init() {
    console_log::init_with_level(Level::Debug).expect("Failed initializing console log");

The main function that handles the queue event should free all memory after it’s done, so isn’t ilkely to cause this.