Minimum min.insync.replicas for Durability

min.insync.replicas must be >= 2 and <= replicationFactor − 1.

Rationale

Producers with `acks=all` are only safe when `min.insync.replicas >= 2`. At the default of 1, a single ISR can ack a write and a broker crash erases data the producer thought was durable. Setting it equal to RF kills availability: any broker loss makes the topic write-unavailable. Sweet spot is `2 <= min.insync.replicas <= RF - 1`.

Pattern

2 <= min.insync.replicas <= replicationFactor - 1

Examples

RF=3 + min.insync.replicas=2

RF=5 + min.insync.replicas=3

RF=3 + min.insync.replicas=1 (silent data loss)

RF=3 + min.insync.replicas=3 (any broker loss = write-unavailable)

Parameters

Name	Default	Description
`min`	`2`	Floor for min.insync.replicas.

Governs

This policy relates to the following Kafka configuration keys:

Implementation

Drop this YAML into Conduktor Console as a ResourcePolicy, then link it from an ApplicationInstance, Application, or KafkaCluster.

Conduktor ResourcePolicy

# Conduktor self-service ResourcePolicy
# spec.configs values are strings — wrap with int(string(...)) for numeric compare.
---
apiVersion: self-serve/v1
kind: ResourcePolicy
metadata:
  name: min-insync-replicas-bound
spec:
  targetKind: Topic
  description: min.insync.replicas must be >= 2 and <= replicationFactor - 1
  rules:
    - condition: '"min.insync.replicas" in spec.configs && int(string(spec.configs["min.insync.replicas"])) >= 2 && int(string(spec.configs["min.insync.replicas"])) <= int(spec.replicationFactor) - 1'
      errorMessage: "min.insync.replicas must be >= 2 and <= replicationFactor - 1 (durability vs availability tradeoff)"

Related policies

Try Conduktor Console

Enforce policies like this across your team — central audit history, pre-commit guardrails, ApplicationInstance bindings. 5-min Docker install.

Get Started →