Module: Lich::Common::BestEffortShutdownCleanup

Defined in:
documented/common/best_effort_shutdown_cleanup.rb

Overview

Performs local cleanup after the frontend or game connection is disrupted.

Unlike OrderlyShutdown, this runner does not assume game request/response I/O is still usable. It preserves local state and runs bounded script teardown as best effort, then lets the existing post-join closeout handle sockets, databases, lifecycle unregister, and process exit.

Defined Under Namespace

Classes: Result

Class Method Summary collapse

Class Method Details

.run(coordinator:, initial_scripts:, remaining_scripts:, script_drain:, vars:, active_sessions_lifecycle: nil, slow_threshold: 1.5) ⇒ Result

Executes best-effort local cleanup once.

This path is intended for connection disruption after critical game I/O has already stopped or become unreliable. It avoids closing the game socket itself and does not describe script hooks as graceful.

Parameters:

  • coordinator (Lich::Common::ShutdownCoordinator) —

    state/intent coordinator

  • initial_scripts (Array<#kill,#name>) —

    scripts present at cleanup start

  • remaining_scripts (#call) —

    returns scripts still registered

  • script_drain (#run) —

    script drain service

  • vars (#save) —

    local script settings persistence

  • active_sessions_lifecycle (#update_connected, nil) (defaults to: nil) —

    optional session registry

  • slow_threshold (Float) (defaults to: 1.5) —

    seconds before script drain reports slow scripts

Returns:

  • (Result) —

    stored best-effort cleanup result

Raises:

  • (ArgumentError) —

    when caller input is invalid



53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
 
# File 'documented/common/best_effort_shutdown_cleanup.rb', line 53

def run(coordinator:, initial_scripts:, remaining_scripts:, script_drain:, vars:, active_sessions_lifecycle: nil, slow_threshold: 1.5)
  validate!(coordinator: coordinator, remaining_scripts: remaining_scripts, script_drain: script_drain, vars: vars)

  result = Result.new(
    completed: false,
    failures: [],
    scripts_drained: false,
    vars_saved: false
  )

  stored_result = coordinator.begin_best_effort_cleanup(result)
  return stored_result unless stored_result.equal?(result)

  log_info("best-effort shutdown cleanup starting reason=#{coordinator.reason || :unknown}")
  update_active_sessions(result, active_sessions_lifecycle)
  drain_scripts(result, initial_scripts, remaining_scripts, script_drain, slow_threshold)
  save_vars(result, vars)
  finish(result)
end