Docs What we monitor
What we monitor
Every check, where its data comes from, how fresh it is, and what it can't see.
Each fully connected client environment is checked every 30 minutes. One check makes at most about 10 API requests to the client org, which is under 500 a day. Every check reports its own result, so one failure never hides the others. A check that fails is shown as failed, never as “nothing wrong”.
Checks
| Check | Source in the client org | What you see |
|---|---|---|
| Connection | The integration user’s identity | That the connection works, and that the org ID is the one you set up |
| Limits | The REST API limits resource | Every limit the org reports with an allowance, closest to its maximum first: daily API requests, data and file storage, async Apex and others |
| Failed logins | Login history, last 24 hours | Failures by the reason Salesforce gives, by user and by hour |
| Failed Flows | Failed Flow interviews | Each failing Flow with its version, element and error, linked to the Flow in the client’s Setup |
| Apex jobs | Async Apex jobs, last 7 days | Batch, queueable, scheduled and future jobs that failed or reported errors |
| Bulk jobs | Bulk API 2.0 ingest jobs | Jobs that failed, with the object and operation. Never the records. |
| Security health | Health Check | The score and the high-risk settings |
| Setup changes | Setup Audit Trail, last 7 days | Who changed what in Setup |
| Scheduled jobs | Scheduled jobs and the async queue | What’s scheduled, what’s queued, and paused Flow interviews |
| Users and licences | User licences and users | Licence use, users who haven’t logged in for 90 days, frozen or locked users |
| Automation | Active Flows and workflow rules | How much automation runs, by type |
Flow error emails come in as they happen, within seconds. For Flow errors only environments they are the only source. See Flow errors only.
Freshness
- Each panel shows when its data was collected.
- When a check fails, it says so in amber and shows the last good data with its time.
- For Flow errors only environments, the time of the last email received takes the place of the collection time.
What we can’t see
- Flow failures that a fault path handles. If the Flow catches its own error, Salesforce may not save a failed interview or send an error email. Its documentation doesn’t say for certain.
- Failed interviews beyond Salesforce’s caps. Salesforce saves at most 100 failed interviews per Flow and 3,000 per org in 24 hours, and none over 1 MB.
- Integration errors outside Bulk API and Apex jobs, such as failed REST calls from another system or failed outbound messages. Salesforce exposes those only to orgs with the Event Monitoring add-on, which we don’t use.
- Real-time login events. Those also need Event Monitoring. We read login history every 30 minutes.
- Unhandled Apex exception emails. They can reach the same address, but are ignored for now.