Polling of Maintenance Mode hosts beyond a deleted network device

Products

• Network Observability

Issue/Introduction

We have received a report stating that when Network Device A was deleted from the OneClick topology, "Reachability Restored" alarms were generated for Host B and Host C, even though they were supposed to be in Maintenance Mode. 

Here are the questions:

When a device on the communication path is deleted, does the system initiate polling for the devices located beyond it?
In that scenario, is it expected behavior for alarms to be generated from devices that are currently in Maintenance Mode?

Environment

Spectrum - 24.3.x/25.4.x

OS - Windows/Linux

Resolution

1. Does the system initiate polling for devices located beyond a deleted device?

Yes. When you delete a device (Network Device A) that is positioned between the SpectroSserver and other hosts (Host B and C), Spectrum’s Fault Isolation logic detects a change in the topology path.

• Path Re-evaluation: Spectrum must determine how to reach Host B and C now that the primary "hop" (Device A) is gone.
• Polling Initiation: To verify the state and connectivity of the downstream hosts via any alternative paths, Spectrum will initiate an immediate poll (ICMP or SNMP) to those devices. This happens regardless of their scheduled polling interval to ensure the topology map remains accurate.

2. Is it expected behavior for alarms to be generated from devices in Maintenance Mode?

Technically, yes, in this specific scenario. While Maintenance Mode is designed to suppress "Contact Lost" alarms, the "Reachability Restored" alarm is a Clear/Informational event that can be triggered by the re-evaluation process.

• The "Reachability Restored" Logic: If Host B and C were previously marked as "Suppressed" or "Unreachable" because of Device A's status, deleting Device A causes Spectrum to successfully poll them directly. A successful poll on a device that was previously "Unreachable" (due to its parent being down or missing) results in a "Reachability Restored" event.
• Maintenance Mode Behavior: Maintenance Mode primarily suppresses Critical (Red) and Major (Orange) alarms. It does not always suppress "Restored" or "Clear" events because these are considered status updates rather than fault alerts.
• Visibility: By default, "Reachability Restored" (Event ID 0x10d00) is often configured to clear existing faults. If the devices were in a "Suppressed" state before being deleted, the restoration event will fire to update their status to "Maintenance" (Brown) from "Unreachable."