From 59ef4fa9d7bef780b0191ab44b220671d7670c75 Mon Sep 17 00:00:00 2001 From: Edwinhr716 Date: Sat, 14 Jun 2025 21:42:18 +0000 Subject: [PATCH] added types of podManagementPolicy --- .../concepts/workloads/controllers/statefulset.md | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/content/en/docs/concepts/workloads/controllers/statefulset.md b/content/en/docs/concepts/workloads/controllers/statefulset.md index 3a79abb5d6841..56c1a1ba7531d 100644 --- a/content/en/docs/concepts/workloads/controllers/statefulset.md +++ b/content/en/docs/concepts/workloads/controllers/statefulset.md @@ -363,6 +363,18 @@ that are running with the `MaxUnavailableStatefulSet` enabled. {{< /note >}} +#### PodManagementPolicy + +You can determine the order at which the pods are deleted by specifying the `.spec.PodManagementPolicy` +field. There are two policies that you can configure for each StatefulSet: + +`parallel` +: When updating, there will always be maxUnavailable number of pods terminating except the last batch + +`orderedReady` +: When updating, the number of pods terminating won't always be maxUnavailable, but sometimes less than + that. This implementation avoids out of order Terminations of pods. + ### Forced rollback When using [Rolling Updates](#rolling-updates) with the default