Saw a crash with this stack trace

[athas]

Traceback (most recent call last):
  File "/usr/local/lib/python3.9/dist-packages/flask/app.py", line 2528, in wsgi_app
    response = self.full_dispatch_request()
  File "/usr/local/lib/python3.9/dist-packages/flask/app.py", line 1825, in full_dispatch_request
    rv = self.handle_user_exception(e)
  File "/usr/local/lib/python3.9/dist-packages/flask/app.py", line 1823, in full_dispatch_request
    rv = self.dispatch_request()
  File "/usr/local/lib/python3.9/dist-packages/flask/app.py", line 1799, in dispatch_request
    return self.ensure_sync(self.view_functions[rule.endpoint])(**view_args)
  File "/views.py", line 99, in index
    result = sessions.create_session()
  File "/sessions.py", line 104, in create_session
    session = Session(
  File "/session.py", line 197, in __init__
    self.process = FutharkREPL(
  File "/session.py", line 115, in __init__
    self.banner, self.init_lastline = next(self.generator).rsplit("\n", 1)
  File "/session.py", line 141, in create_generator
    yield self.read()
  File "/session.py", line 123, in read
    out = timed_read(
  File "/session.py", line 46, in timed_read
    ready_to_read, _, _ = select.select(fds, [], [], step_timeout)
ValueError: filedescriptor out of range in select()
    r, w, e = select.select(r, w, e, timeout)
ValueError: filedescriptor out of range in select()

I'll look at it eventually; probably a missing exception handler in some cases of network communication failure. Alternatively we could set up the REPL daemon to restart on crash.