Kubernetes is preparing a significant change for Linux cluster operators: from version 1.35, the failCgroupV1 setting will default to true, according to Kubernetes’ guidance. Administrators still using cgroup v1 will need to migrate nodes and configure suitable cgroup drivers before upgrading.
Watch Desk analysis
What happened
Kubernetes’ 6 October guidance describes the move from cgroup v1 to cgroup v2 and its practical implications for upgrades. It says the setting that fails on cgroup v1 will default to true in Kubernetes 1.35, making the migration relevant to operators planning their next upgrade.
The change also brings cgroup v2 capabilities including unified resource management, container-level OOM group killing and PSI metrics. The guidance says it supports Memory QoS and in-place vertical scaling, too. Read Kubernetes’ cgroup v2 guidance.
Why it matters
This is an upgrade-planning issue, not just a change in terminology. Operators who have not moved Linux nodes to cgroup v2, or who have not configured the relevant cgroup drivers, may need to do that work before upgrading. The new default gives teams a clear reason to check their node configuration rather than discover a mismatch during the upgrade itself.
The transition also brings resource-management and monitoring capabilities, but the immediate reader takeaway is straightforward: check whether your Linux nodes use cgroup v1 and plan the migration before Kubernetes 1.35.
Our read
A default change can be easy to overlook until it becomes a very unfun upgrade surprise. Cluster operators should treat cgroup v2 readiness as a preparation task, and consult the project’s guidance for the details relevant to their setup.
What to watch
- When Kubernetes 1.35 is released and how the default behaves in practice.
- Whether the project publishes further migration guidance for node and cgroup-driver configurations.
Discussion spark: Should Kubernetes make cgroup v2 the default even if that forces lagging clusters to plan a migration, or should the project keep a softer transition for longer?
Sources and evidence
- The Shift to cgroup v2 in Kubernetes: What You Need to Know (6 October 2026, 18:00 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.