Skip to content

PAUSE_RECOVERY.md

Latest commit

 

History

History
60 lines (45 loc) · 2.26 KB

PAUSE_RECOVERY.md

File metadata and controls

60 lines (45 loc) · 2.26 KB
Raw

Admin pause and recovery behavior

VaultQuest pause controls are for incident response: protect user funds, preserve readable state, and give frontend/backend services a predictable mode while maintainers recover.

Contract expectations

When paused, read-only methods remain available. State-changing user actions return a paused error before moving funds or changing pool state.

Action Paused behavior
create Blocked
join Blocked
drip Blocked
claim Blocked unless maintainers explicitly enable emergency claims
withdraw Blocked by default; emergency withdrawals are a follow-up contract feature
admin config changes Allowed only for recovery-safe settings
pause Idempotent; emits paused
recover / unpause Admin-only; emits recovered

Follow-up implementation gaps for the current scaffold:

  • Add persisted paused state and admin authorization to the contract.
  • Add explicit paused errors for create, join, drip, claim, and withdraw.
  • Decide whether emergency withdrawals/claims should be separate admin-enabled modes.

Frontend behavior

During pause, disable create, join, drip, claim, and withdraw controls. Reads, pool detail pages, balances, and reward history stay visible.

Suggested user-facing copy:

  • Banner title: VaultQuest is temporarily paused
  • Body: Pool actions are disabled while maintainers complete recovery. Your balances and history remain visible.
  • Button/tooltips: Action unavailable during pause

After recovery, remove the banner, re-enable actions, and refresh pool and position data.

Backend and indexer behavior

Indexers must continue ingesting events while paused, especially paused, recovered, config_changed, and any late settlement events. Backend reads remain available. Mutation endpoints may accept frontend intents, but should mark or reject paused actions consistently once the contract exposes a canonical paused error.

Operational recovery:

  1. Admin emits paused with a short reason.
  2. Frontend/backend surface paused state and keep reads live.
  3. Maintainers patch configuration or deploy the contract fix.
  4. Admin emits recovered.
  5. Indexer confirms recovered; backend clears paused state; frontend refreshes and re-enables actions.