Agent Status Cascade on Discovery Client Offline
When a Discovery Client went offline, its registered agents previously continued to show as Live because the client-down path never updated the agents' connected flags. This made it hard to diagnose why scheduled scans were not running. The offline status now cascades to all agents owned by the client. On recovery, each agent re-registers itself as Live via its own heartbeat.
Problems and fixes
The following root causes were addressed:
| Problem | Fix |
|---|---|
| When a Discovery Client was marked offline, agents kept their old Live status. | Every agent owned by the client is flipped to Not Live in the same operation. |
| The cascade-on-down was missing; the recovery path was already correct. | A down-cascade was added; the recovery path continues to use the existing per-agent heartbeat. |
Where the fix appears
The corrected status behavior is visible in the following locations:
| Location | What changed |
|---|---|
| Discovery > Agents list | When a Discovery Client goes offline, all of its agents immediately switch to Not Live. |
| Audit log | Client offline state is consistent with the agent list, making scan diagnosis easier. |
Note: No action is required. After the patched server release, agent live and offline status reflects client connectivity automatically.