Nm state sync

[cyc60]

After the global state is updated (every 24 hours), the operator must sync their individual state:

1. Watch for StateUpdated events or poll GET /nodes-manager/state-vote to detect new state updates
2. Fetch the state data from IPFS (contains operator's totalAssets, cumPenaltyAssets, cumEarnedFeeShares, and merkle proof)
3. If vault needs harvesting, call updateVaultState(harvestParams) first
4. Submit NodesManager.updateOperatorState(params) with the merkle proof — can be batched via multicall

This should run as a periodic task. The operator must stay synced to the latest nonce — it's a prerequisite for entering the exit queue and claiming assets.

[Copilot]

Pull request overview

Adds an operator-side “state sync” flow to keep an individual operator’s on-chain state aligned with the latest globally-published NodesManager state (fetched from IPFS), optionally batching a vault harvest via multicall.

Changes:

• Introduces StateSyncTask to detect when the operator’s nonce lags the global nonce and submit updateOperatorState (optionally with updateVaultState).
• Adds IPFS fetch + tx submission helpers (fetch_operator_state_from_ipfs, submit_state_sync_transaction) and a new params typing.
• Extends NodesManager contract wrapper/encoder to support querying state update events and encoding operator/vault update calls.

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
src/node_manager/typings.py	Adds OperatorStateUpdateParams dataclass for updateOperatorState.
src/node_manager/tasks.py	Adds StateSyncTask periodic loop that detects new global state and syncs the operator.
src/node_manager/execution.py	New helper module to fetch operator params from IPFS and submit the sync transaction (with optional multicall).
src/common/contracts.py	Adds NodesManager event query + nonce helpers, and encoder methods for operator/vault state updates.
src/common/app_state.py	Adds in-memory checkpoint (state_sync_block) for event scanning.
src/commands/node_manager_start.py	Runs the new StateSyncTask alongside the existing NodeManagerTask.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

ipfs_fetch_client.fetch_json() is typed to return dict | list, but this code assumes a dict and calls .get('operators', ...). If IPFS returns a JSON array (or otherwise unexpected shape), this will raise at runtime. Consider validating ipfs_data is a dict with an operators key (and logging/returning None on unexpected data) before iterating.

Suggested change

	for operator_data in ipfs_data.get('operators', []): # type: ignore
	if not isinstance(ipfs_data, dict):
	logger.error(
	"Unexpected IPFS data type for hash %s: expected dict, got %s",
	ipfs_hash,
	type(ipfs_data).__name__,
	)
	return None
	operators = ipfs_data.get('operators')
	if not isinstance(operators, list):
	logger.error(
	"Missing or invalid 'operators' field in IPFS data for hash %s",
	ipfs_hash,
	)
	return None
	for operator_data in operators:

[Copilot]

The contract ABI expects totalAssets, cumPenaltyAssets, and cumEarnedFeeShares as uint128 and the proof as bytes32[], but the values parsed from IPFS are not validated. A negative/oversized int or a proof element not exactly 32 bytes will lead to an on-chain revert (wasting gas). Add explicit validation (0 <= value < 2**128, len(proof_item)==32) and fail fast with a clear log message when the IPFS payload is malformed.

[Copilot]

This PR introduces a new periodic task (StateSyncTask) and IPFS parsing/transaction submission logic, but there are no accompanying tests. The repo already has task-level tests for other modules (e.g. src/meta_vault/tests/test_tasks.py, src/withdrawals/tests/test_tasks.py). Consider adding unit tests that mock NodesManagerContract, ipfs_fetch_client, and transaction_gas_wrapper to cover: already-synced path, missing operator in IPFS, harvest+multicall path, and failed receipt status.