1
0
mirror of https://github.com/openshift/openshift-docs.git synced 2026-02-05 12:46:18 +01:00

Documentation is unclear about status of cgroups v2

This commit is contained in:
Michael Burke
2022-12-12 15:10:55 -05:00
committed by openshift-cherrypick-robot
parent 6aed9edd55
commit db7230510f
6 changed files with 214 additions and 2 deletions

View File

@@ -489,6 +489,7 @@ Topics:
File: configuring-firewall
- Name: Enabling Linux control group version 2 (cgroup v2)
File: enabling-cgroup-v2
Distros: openshift-enterprise
- Name: Validating an installation
File: validating-an-installation
Distros: openshift-origin,openshift-enterprise
@@ -2161,7 +2162,10 @@ Topics:
Distros: openshift-enterprise,openshift-origin
- Name: Enabling Linux control group version 2 (cgroup v2)
File: nodes-cluster-cgroups-2
Distros: openshift-enterprise,openshift-origin
Distros: openshift-enterprise
- Name: Configuring the Linux cgroup on your nodes
File: nodes-cluster-cgroups-okd
Distros: openshift-origin
- Name: Enabling features using FeatureGates
File: nodes-cluster-enabling-features
Distros: openshift-enterprise,openshift-origin

View File

@@ -0,0 +1,126 @@
// Module included in the following assemblies:
//
// * nodes/clusters/nodes-cluster-cgroups-okd.adoc
// * post_installation_configuration/cluster-tasks.adoc
ifeval::["{context}" == "nodes-cluster-cgroups-2"]
:node:
endif::[]
ifeval::["{context}" == "post-install-cluster-tasks"]
:post:
endif::[]
:_content-type: PROCEDURE
[id="nodes-clusters-cgroups-okd-configure_{context}"]
= Configuring Linux cgroup
ifdef::post[]
By default, {product-title} uses link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v2.html[Linux control group version 2] (cgroup v2) in your cluster. You can switch to link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v1.html[Linux control group version 1] (cgroup v1), if needed.
cgroup v2 is the next version of the kernel link:https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/resource_management_guide/ch01[control group] and offers multiple improvements. However, it can have some unwanted effects on your nodes.
endif::post[]
You can configure whether your cluster uses cgroup v1 or cgroup v2 by editing the `node.config` object. Enabling the other version of cgroup in {product-title} disables the current cgroup controllers and hierarchies in your cluster.
.Prerequisites
* Have administrative privilege to a working {product-title} cluster.
.Procedure
. Create a `MachineConfig` object file that identifies the kernel argument (for example, `worker-cgroup-v1.yaml`)
+
[source,yaml]
----
apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
labels:
machineconfiguration.openshift.io/role: worker <1>
name: worker-cgroup-v1 <2>
spec:
config:
ignition:
version: 3.2.0
kernelArguments:
- systemd.unified_cgroup_hierarchy=0 <3>
----
+
<1> Applies the new kernel argument only to worker nodes.
<2> Applies a name to the machine config.
<3> Configures cgroup v1 on the associated nodes.
. Create the new machine config:
+
[source,terminal]
----
$ oc create -f 05-worker-cgroup-v1.yaml
----
. Check to see that the new machine config was added:
+
[source,terminal]
----
$ oc get MachineConfig
----
+
.Example output
[source,terminal]
----
NAME GENERATEDBYCONTROLLER IGNITIONVERSION AGE
00-master 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
00-worker 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
01-master-container-runtime 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
01-master-kubelet 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
01-worker-container-runtime 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
01-worker-kubelet 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
99-worker-cgroup-v1 3.2.0 105s
99-master-generated-registries 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
99-master-ssh 3.2.0 40m
99-worker-generated-registries 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
99-worker-ssh 3.2.0 40m
rendered-master-23e785de7587df95a4b517e0647e5ab7 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
rendered-master-c5e92d98103061c4818cfcefcf462770 60746a843e7ef8855ae00f2ffcb655c53e0e8296 3.2.0 115s
rendered-worker-5d596d9293ca3ea80c896a1191735bb1 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
----
. Check the nodes:
+
[source,terminal]
----
$ oc get nodes
----
+
.Example output
[source,terminal]
----
NAME STATUS ROLES AGE VERSION
ip-10-0-136-161.ec2.internal Ready worker 28m v1.25.0
ip-10-0-136-243.ec2.internal Ready master 34m v1.25.0
ip-10-0-141-105.ec2.internal Ready,SchedulingDisabled worker 28m v1.25.0
ip-10-0-142-249.ec2.internal Ready master 34m v1.25.0
ip-10-0-153-11.ec2.internal Ready worker 28m v1.25.0
ip-10-0-153-150.ec2.internal Ready master 34m v1.25.0
----
+
You can see that The command disables scheduling on each worker node.
. Check that the sys/fs/cgroup/cgroup2fs file has been moved to the `tmpfs` file system:
+
[source,terminal]
----
$ stat -c %T -f /sys/fs/cgroup
----
+
.Example output
+
[source,terminal]
----
tmpfs
----
ifeval::["{context}" == "nodes-cluster-cgroups-2"]
:!node:
endif::[]
ifeval::["{context}" == "post-install-cluster-tasks"]
:!post:
endif::[]

View File

@@ -20,7 +20,30 @@ Examples of kernel arguments you could set include:
* **nosmt**: Disables symmetric multithreading (SMT) in the kernel. Multithreading allows multiple logical threads for each CPU. You could consider `nosmt` in multi-tenant environments to reduce risks from potential cross-thread attacks. By disabling SMT, you essentially choose security over performance.
ifndef::openshift-origin[]
* **systemd.unified_cgroup_hierarchy**: Enables link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v2.html[Linux control group version 2] (cgroup v2). cgroup v2 is the next version of the kernel link:https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/resource_management_guide/ch01[control group] and offers multiple improvements.
+
[IMPORTANT]
====
[subs="attributes+"]
{product-title} cgroups version 2 support is a Technology Preview feature only. Technology Preview features
are not supported with Red Hat production service level agreements (SLAs) and
might not be functionally complete. Red Hat does not support Technology Preview features for use in production systems. These features provide early access to upcoming product
features, enabling customers to test functionality and provide feedback during
the development process.
For more information about the support scope of Red Hat Technology Preview
features, see https://access.redhat.com/support/offerings/techpreview/.
====
endif::openshift-origin[]
ifdef::openshift-origin[]
* **systemd.unified_cgroup_hierarchy**: Configures the version of Linux control group that is installed on your nodes: link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v1.html[cgroup v1] or link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v2.html[cgroup v2]. cgroup v2 is the next version of the kernel link:https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/resource_management_guide/ch01[control group] and offers multiple improvements. However, it can have some unwanted effects on your nodes.
+
[NOTE]
====
cgroup v2 is enabled by default. To disable cgroup v2, use the `systemd.unified_cgroup_hierarchy=0` kernel argument, as shown in the following procedure.
====
endif::openshift-origin[]
See link:https://www.kernel.org/doc/Documentation/admin-guide/kernel-parameters.txt[Kernel.org kernel parameters] for a list and descriptions of kernel arguments.
@@ -61,6 +84,7 @@ rendered-master-23e785de7587df95a4b517e0647e5ab7 52dd3ba6a9a527fc3ab42afac8d12
rendered-worker-5d596d9293ca3ea80c896a1191735bb1 52dd3ba6a9a527fc3ab42afac8d12b693534c8c9 3.2.0 33m
----
ifndef::openshift-origin[]
. Create a `MachineConfig` object file that identifies the kernel argument (for example, `05-worker-kernelarg-selinuxpermissive.yaml`)
+
[source,yaml]
@@ -83,6 +107,38 @@ spec:
<2> Named to identify where it fits among the machine configs (05) and what it does (adds
a kernel argument to configure SELinux permissive mode).
<3> Identifies the exact kernel argument as `enforcing=0`.
. Create the new machine config:
+
[source,terminal]
----
$ oc create -f 05-worker-kernelarg-selinuxpermissive.yaml
----
endif::openshift-origin[]
ifdef::openshift-origin[]
. Create a `MachineConfig` object file that identifies the kernel argument (for example, `05-worker-kernelarg-selinuxpermissive.yaml`)
+
[source,yaml]
----
apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
labels:
machineconfiguration.openshift.io/role: worker <1>
name: 05-worker-kernelarg-selinuxpermissive <2>
spec:
config:
ignition:
version: 3.2.0
kernelArguments:
- enforcing=0 <3>
systemd.unified_cgroup_hierarchy=0 <4>
----
+
<1> Applies the new kernel argument only to worker nodes.
<2> Named to identify where it fits among the machine configs (05) and what it does (adds
a kernel argument to configure SELinux permissive mode).
<3> Identifies the exact kernel argument as `enforcing=0`.
<4> Configures cgroup v1 on the associated nodes. cgroup v2 is the default.
. Create the new machine config:
+
@@ -90,6 +146,7 @@ a kernel argument to configure SELinux permissive mode).
----
$ oc create -f 05-worker-kernelarg-selinuxpermissive.yaml
----
endif::openshift-origin[]
. Check the machine configs to see that the new one was added:
+

View File

@@ -0,0 +1,19 @@
:_content-type: ASSEMBLY
:context: nodes-cluster-cgroups-2
[id="nodes-cluster-cgroups-okd"]
= Configuring the Linux cgroup on your nodes
include::_attributes/common-attributes.adoc[]
toc::[]
By default, {product-title} uses link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v2.html[Linux control group version 2] (cgroup v2) in your cluster. You can switch to link:https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v1.html[Linux control group version 1] (cgroup v1), if needed.
cgroup v2 is the next version of the kernel link:https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/6/html/resource_management_guide/ch01[control group] and offers multiple improvements. However, it can have some unwanted effects on your nodes.
// The following include statements pull in the module files that comprise
// the assembly. Include any combination of concept, procedure, or reference
// modules required to cover the user story. You can also include other
// assemblies.
include::modules/nodes-clusters-cgroups-okd-configure.adoc[leveloffset=+1]

View File

@@ -587,8 +587,12 @@ include::modules/machine-autoscaler-cr.adoc[leveloffset=+2]
:FeatureResourceName: MachineAutoscaler
include::modules/deploying-resource.adoc[leveloffset=+2]
ifndef::openshift-origin[]
include::modules/nodes-clusters-cgroups-2.adoc[leveloffset=+1]
endif::openshift-origin[]
ifdef::openshift-origin[]
include::modules/nodes-clusters-cgroups-okd-configure.adoc[leveloffset=+1]
endif::openshift-origin[]
[id="post-install-tp-tasks"]
== Enabling Technology Preview features using FeatureGates

View File

@@ -74,3 +74,5 @@ Besides managing `MachineConfig` objects, the MCO manages two custom resources (
include::modules/create-a-kubeletconfig-crd-to-edit-kubelet-parameters.adoc[leveloffset=+2]
include::modules/create-a-containerruntimeconfig-crd.adoc[leveloffset=+2]
include::modules/set-the-default-max-container-root-partition-size-for-overlay-with-crio.adoc[leveloffset=+2]