Monitoring deliveries

Dashboard

The Dashboard summarizes recent delivery health over a rolling 24-hour window, each compared to the prior 24 hours:

• Success rate and error rate across deliveries.
• p95 latency — how long your endpoints take to respond.
• Deliveries (24h) and lifetime total delivered.
• A throughput chart of successes vs. failures over time.

Per-webhook metrics

The Webhooks page rolls up live numbers for each endpoint — events per second, p95 latency, success rate, and current health state — so you can spot a single misbehaving destination at a glance.

Event logs

The Logs page lists individual events. Filter by object, change type, status, webhook, and date range to narrow down what you're looking for.

Open any event to see its full detail:

• The event payload and which fields changed.
• Each delivery attempt in order — attempt number, HTTP status, response body, error message, and duration in milliseconds.

A delivery's attempt history is the fastest way to diagnose why an endpoint rejected an event (for example a 401 from a missing header or a 500 from the receiver).

Replaying from the logs

From an event's detail you can re-enqueue it for delivery — handy after fixing a receiver. See Delivery, retries & DLQ for how replay and the dead-letter queue work together.

Stream gaps (possible missed events)

SFHooks tracks its exact position in your org's Change Data Capture stream (the replay id) so that after any disconnection it resumes precisely where it left off. Salesforce retains events for about 72 hours — if a connection is down longer than that, the saved position expires and the stream can only resume from the latest event. Events published during the outage can no longer be replayed.

When that happens, SFHooks records a stream gap and shows a “Possible missed events” indicator on the Connections page. Click it to see each incident's window — the span between the stream's last confirmed position and the moment it reconnected. Review records changed in that window and re-sync them in your downstream systems if needed; delivery is current from the moment of reconnection. On plans with Health Alerts, the org's owners and admins are also emailed when a gap is recorded.

Usage

The Usage page shows your current billing-period event count against your plan limit, the features available on your plan, and any trial end date. Remember that logs are only retained for your plan's window — export or act on anything you need beyond it.