CTF-703: Event Auto-Cleanup¶
Statement¶
The CTF layer should automatically clean up event resources after the event transitions to ended or cancelled state. Cleanup shall orchestrate Engine range destruction via cms.services.destroy_range() for all participant ranges. Cleanup shall be configurable with a delay (for example 2 hours after event end) to allow post-event review, or immediate on cancellation. Session management is a platform auth concern and not handled by CTF cleanup. The system shall log all cleanup actions.
Rationale¶
Forgotten range instances after events are the primary source of unnecessary AWS costs in Shifter. Automatic cleanup ensures cloud resources are released even if organizers forget. CTFd does not manage infrastructure, but Shifter must handle this because each participant has dedicated cloud resources that cost money every minute they run.
Traceability¶
- IMPLEMENTS → CODE_FILE
shifter/shifter_platform/ctf/models/(CTF models - CTFEvent (auto_cleanup, cleanup_delay_hours, get_cleanup_time), CTFScheduledTask) - IMPLEMENTS → CODE_FILE
shifter/shifter_platform/ctf/services/event/(CTF Event service - _schedule_event_tasks() schedules CLEANUP_RANGES with configurable delay) - IMPLEMENTS → CODE_FILE
shifter/shifter_platform/ctf/bridges.py(CTF bridges - cms_destroy_range() cross-domain integration for range teardown) - TESTS → TEST
shifter/shifter_platform/tests/ctf/test_models.py(Model tests - test_event_get_cleanup_time, auto_cleanup defaults) - IMPLEMENTS → CODE_FILE
shifter/shifter_platform/ctf/services/range/(CTF Range service - cleanup_event_ranges(), _destroy_single_range()) - IMPLEMENTS → CODE_FILE
shifter/shifter_platform/ctf/management/commands/run_ctf_scheduler.py(CTF scheduler - _handle_cleanup_ranges() dispatches cleanup_event_ranges()) - TESTS → TEST
shifter/shifter_platform/tests/ctf/test_services/test_range.py(Range service tests - TestCleanupEventRanges, TestDestroyParticipantRange) - DOCUMENTS → GITHUB_ISSUE
Brad-Edwards/bigrae#660(CTF-703: Event Auto-Cleanup)