AHG Central — Fleet Monitoring (AtoM Heratio)

AHG Central is the optional cloud service from The Archive and Heritage Group that lets a group of AtoM Heratio installs be watched from one place. Each install reports a daily heartbeat (it is alive, and on which version) and — when the operator opts in — an hourly error-log sync. The fleet is monitored from the AHG Workbench.

AHG Central is advisory. If it is switched off, unreachable, or the install is air-gapped, AtoM Heratio runs exactly as before — nothing in the product depends on it.

Configure it at Admin → AHG Settings → AHG Integration.

Onboarding is automatic

A fresh AtoM Heratio install joins the fleet without any registration form:

• the AHG Central client ships inside the AHG plugin collection (ahgCorePlugin's AhgCentralService, with the optional standalone ahgCentralPlugin as a universal alternative);
• the install reads its connection settings from the AHG Integration page;
• it identifies itself from the server hostname (atom-<hostname>);
• on its first heartbeat, AHG Central registers it automatically.

Paste the fleet enrolment key once on the AHG Integration page, tick Enable AHG Central Integration, and save. The Test Connection button confirms reachability before any data flows.

Cron is required

Unlike the Heratio Laravel platform — which auto-schedules the heartbeat — the AtoM Heratio platform ships the tasks but does not auto-schedule them. Each install needs an explicit cron entry. The AHG Integration page shows the exact commands; the default pattern is:

0 5 * * *   cd /path/to/atom && php symfony central:heartbeat
30 * * * *  cd /path/to/atom && php symfony central:sync-errors

(If the standalone ahgCentralPlugin is in use, the task names are ahg:central-heartbeat and ahg:central-sync-errors — schedule one heartbeat path, not both.)

Error-log sync (opt-in)

Sync Error Logs to AHG Central is a separate switch, off by default. Error logs can contain sensitive detail, so shipping them off the server needs a deliberate decision.

When you turn it on, AtoM Heratio sends the current set of open errors from the ahg_error_log table once an hour, redacted before they leave the server: email addresses and long number sequences are masked, and URL query strings are stripped. Stack traces, client IP addresses, user agents and user ids are never sent at all. Only turn this on when your institution is comfortable with off-site error visibility.

Where you watch the fleet

Fleet monitoring lives in the AHG Workbench admin area. There an operator sees every connected install — its version, when it was last seen, whether it is online or has gone quiet — and recent errors across the whole fleet. The Workbench also raises a notification when a new install joins, when an install reports critical errors, or when an install goes silent.

A simple fallback dashboard is available directly at central.theahg.co.za.

The settings

Use the Test Connection button to confirm the install can reach AHG Central before saving.