Soroban Smart Contract Monitoring: What Developers Should Track in Production

You deployed your Soroban smart contract to mainnet. It is live, processing transactions, managing state. But how do you know if it is working correctly? How do you find out when something goes wrong before your users do?

What Can Go Wrong

Failed invocations: Transactions that call your contract but revert

Unexpected storage changes: State modifications that do not match expected patterns

Gas spikes: Invocations consuming abnormally high resources

Event anomalies: Missing events or unexpected event patterns

Storage bloat: Contract storage growing beyond expected bounds

Core Metrics to Track

1. Invocation Success Rate

The most fundamental metric. A healthy contract should have a 95%+ success rate. Failures below that indicate a bug, UI issue, or configuration problem.

2. Gas Consumption

Track average and P95 gas per invocation. A sudden spike usually indicates an inefficient code path, unexpectedly large storage entries, or a new usage pattern.

3. Error Types

Error Type	Meaning
`InvokeHostFunctionFailed`	Contract execution reverted
`InsufficientBalance`	Caller lacks XLM for fees
`ExceededResourceLimit`	Transaction hit CPU/memory limits
`StorageEntryNotFound`	Missing storage entry

4. Storage State

Track persistent, temporary, and instance storage entry counts and sizes. Storage has direct cost implications — entries that expire must be restored at additional cost.

5. Event Volume

If your contract emits events, monitor volume over time, events by topic, and gaps where events should be firing but are not.

Setting Up Monitoring

Using Stellar RPC Events

Poll the getEvents RPC method at regular intervals:

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "getEvents",
  "params": {
    "startLedger": 61000000,
    "filters": [{
      "type": "contract",
      "contractIds": ["YOUR_CONTRACT_ID"],
      "topics": [["*"]]
    }],
    "pagination": { "limit": 100 }
  }
}

Alerting Rules

Alert	Condition	Severity
Success rate drop	Below 90% over 1 hour	Critical
No invocations	Zero calls in 30 min (if normally active)	Warning
Gas spike	P95 gas 2x baseline	Warning
Storage growth	10%+ increase in 24 hours	Info
Error burst	5+ failures in 5 minutes	Critical

Common Production Issues

TTL Expiration

Soroban storage entries have time-to-live values. If an entry expires, it must be restored before it can be read, causing unexpected failures. Monitor TTL values for critical entries.

Frontend Staleness

If your frontend caches contract state, users may submit transactions based on stale data. Monitor the delta between frontend state and actual contract state.

Tools

LumenQuery provides contract monitoring tools:

[Smart Contract Explorer](/contracts): Decoded call history, storage viewer, event stream

[Network Analytics](/analytics): Network-level Soroban activity metrics

[Live Transactions](/dashboard/transactions): Real-time transaction stream with Soroban highlighting

*Monitor your Soroban contracts in production with LumenQuery. Decoded calls, storage viewer, and event streaming — start free.*