# Require both Service Fabric balancing thresholds for the same metric

> Why can visibly uneven Service Fabric node load fail to trigger a balancing run?

- Canonical URL: https://update.dsesecurity.com/updates/dse-20260909-134-require-both-service-fabric-balancing-thresholds-for-the-same-metric/
- Publisher: Detection Systems & Engineering (DSE Security)
- Author: DSE Security Editorial Team
- Published: 2026-09-10T00:29:42+00:00
- Modified: 2026-09-10T00:52:39+00:00
- Last reviewed by DSE: 2026-09-09
- Resource type: Guide
- DSE priority: Information
- Topics: Business Continuity, IT
- Reading time: 2 minutes

## What you need to know

Why can visibly uneven Service Fabric node load fail to trigger a balancing run?

## Potentially affected

Service Fabric Cluster Resource Manager metrics using load spreading rather than defragmentation.

## DSE recommendation

DSE recommends recording the relevant load observations, metric name and threshold values together.

## Article

## Source facts

For a load-spreading metric, Service Fabric compares the highest node load with the lowest as a ratio against its balancing threshold. Balancing also requires the activity threshold to be exceeded for that same metric. A balancing search does not guarantee movement: placement constraints or movement cost can prevent an improvement. Defragmentation metrics use the opposite ratio direction for their trigger. [Microsoft Learn](https://learn.microsoft.com/en-us/azure/service-fabric/service-fabric-cluster-resource-manager-balancing).

## Applicability

Identify the metric’s strategy and effective thresholds before diagnosing an idle balancer. Do not combine one metric’s relative imbalance with another metric’s absolute activity. If per-node-type balancing is enabled, include the applicable overrides in the review instead of assuming only the global settings matter.

## DSE recommendation

DSE recommends recording the relevant load observations, metric name and threshold values together. Decide whether the intended behavior is spreading or consolidation before proposing a tuning change. Avoid lowering a threshold merely because two nodes show different loads; first establish whether both trigger conditions are met and whether an eligible move exists.

## Verification

Compare observations taken for the same metric and balancing scope, then inspect the expected trigger decision in a controlled workload test. Distinguish a scheduled search from an actual replica or instance movement. If the search makes no move, examine constraints and movement cost before declaring the thresholds ineffective. Preserve the configured strategy and resulting placement evidence with any proposed adjustment.

## Official references

[Microsoft Learn](https://learn.microsoft.com/en-us/azure/service-fabric/service-fabric-cluster-resource-manager-balancing). Source retrieved September 9, 2026.

## Primary reference

- Name: Balance your Azure Service Fabric cluster - Azure Service Fabric | Microsoft Learn
- Authority: Microsoft Learn
- URL: https://learn.microsoft.com/en-us/azure/service-fabric/service-fabric-cluster-resource-manager-balancing
- Source publication date: Not stated by the source

## Citation and use

Preferred citation: “Require both Service Fabric balancing thresholds for the same metric,” DSE Security, https://update.dsesecurity.com/updates/dse-20260909-134-require-both-service-fabric-balancing-thresholds-for-the-same-metric/
Publishing principles: https://update.dsesecurity.com/updates/dse-updates-editorial-methodology/
Usage and citation policy: https://update.dsesecurity.com/usage/
Copyright © 2026 Detection Systems & Engineering. All rights reserved.