Azure Kubernetes Fleet Manager Adds Max Allowed Failures for Update Runs
Azure Kubernetes Fleet Manager now includes an optional "Maximum Allowed Failures" setting, currently in public preview. This new capability allows update runs to proceed even when a specified number or percentage of member cluster updates encounter failures. Platform teams can define these failure thresholds at the stage or group level, providing greater control over balancing rollout correctness with update progress. This enables more flexible and resilient update strategies for fleet-managed Kubernetes clusters.
Features (1) ›
- Configure maximum allowed failures for update runs
Azure Kubernetes Fleet Manager introduces an optional setting that allows update runs to continue even if some member cluster updates fail. Platform teams can define failure thresholds at either the stage or group level using fixed counts or percentages, balancing rollout correctness with update progress.
https://azure.microsoft.com/updates?id=567939
Related releases
- Azure AI Foundry model o4-mini 2025-04-16 end-of-life date moved to 2026-11-19 endoflife.date ·
- Azure AI Foundry model o3-pro 2025-06-10 end-of-life date moved to 2026-11-19 endoflife.date ·
- Azure AI Foundry model o3-mini 2025-01-31 end-of-life date moved to 2026-11-19 endoflife.date ·
- Azure AI Foundry model o3-deep-research 2025-06-26 end-of-life date moved to 2026-11-19 endoflife.date ·
- Azure AI Foundry model o3 2025-04-16 end-of-life date moved to 2026-11-19 endoflife.date ·
- Azure AI Foundry model o1-pro 2025-03-19 end-of-life date moved to 2026-11-19 endoflife.date ·