Add support for Rocky Linux 10.1 by pranavcracker · Pull Request #797 · Netcracker/KubeMarine

pranavcracker · 2026-03-18T12:54:25Z

Description

This PR adds support for Rocky Linux 10.1 in Kubemarine and introduces improved handling of kernel upgrades during the system preparation phase.

Background

While validating Kubemarine on Rocky Linux 10.1, an issue was observed during the prepare.system.modprobe stage:

Kernel packages are upgraded during prepare.package_manager.manage_packages
The system continues running the old kernel (no reboot yet)
Required kernel modules (e.g., br_netfilter) are unavailable
This leads to failures in modprobe

Solution

Add support for Rocky Linux 10.1
Kernel Upgrade Detection
Early Reboot Mechanism
Avoid Duplicate Reboots

Changes Introduced

1. Rocky Linux 10.1 Support

Added support for rhel10 OS family where applicable
Ensured compatibility with kernel 6.x and updated package behavior

2. Kernel Upgrade Detection

Detects mismatch between:
- Running kernel (uname -r)
- Latest installed kernel
Implemented for:
- RHEL-based systems (rpm -q kernel-core)
- Debian-based systems (/boot/vmlinuz-*)

3. Early Reboot Mechanism

If kernel upgrade is detected:
- Perform immediate reboot of affected nodes
- Uses reboot_group for precise scope
Ensures subsequent steps (e.g., modprobe, sysctl) operate on the correct kernel

4. Avoid Duplicate Reboots

Introduced cluster.context["early_reboot_done"]
Prevents redundant reboot at cumulative stage

How to apply

Install task/s

Test Cases

TestCase 1
Steps:

Install cluster with Rocky Linux 10.1 OS

Results:

Before	After
Failure	Success

TestCase 2: Kernel upgrade during installation (Rocky Linux 10.1)
Steps:

Prepare a fresh VM with Rocky Linux 10.1
Ensure base image contains older running kernel
Run Kubemarine installation
During prepare.package_manager.manage_packages, allow kernel packages to upgrade
Observe behavior before prepare.system.modprobe

Results:

Before	After
`modprobe` failure due to missing kernel modules	Early reboot triggered and `modprobe` succeeds

Checklist

I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
There is no breaking changes, or migration patch is provided
Integration CI passed
Unit tests. If Yes list of new/changed tests with brief description
There is no merge conflicts

Unit tests

Indicate new or changed unit tests and what they do, if any.

theboringstuff · 2026-03-27T11:56:30Z

kubemarine/procedures/install.py

        manage_custom_packages
    ])
-
+    affected_hosts = system.detect_kernel_upgrade(group)


Can you please help to understand following questions:

Which exactly package causes kernel upgrade? Maybe share install log where we can see kernel upgraded

From which version to which version kernel is upgraded?

During installation of several mandatory packages the kernel is getting upgraded as below -

*** TASK prepare.package_manager.manage_packages *** Running kubemarine.procedures.install.manage_mandatory_packages: Group 'balancer' is requested for usage, but this group does not exist. Group 'balancer' is requested for usage, but this group does not exist. Group 'balancer' is requested for usage, but this group does not exist. Group 'balancer' is requested for usage, but this group does not exist. [install.manage_mandatory_packages] Installing ['conntrack-tools', 'iptables-nft', 'openssl', 'curl', 'policycoreutils-python-utils', 'kmod'] on 'k8s-prpa-aio' [executor.queue] Performing sudo ('yum install -y -d1 --color=never conntrack-tools iptables-nft openssl curl policycoreutils-python-utils kmod; PACKAGES=$(rpm -q conntrack-tools iptables-nft openssl curl policycoreutils-python-utils kmod); if [ $? != 0 ]; then echo "$PACKAGES" | grep \'is not installed\'; echo "Failed to check version for some packages. Make sure packages are not already installed with higher versions. Also, make sure user-defined packages have rpm-compatible names. "; exit 1; fi ',) on nodes ['10.102.0.126'] with options: {'warn': False, 'hide': True, 'pty': True, 'env': None, 'timeout': None} [executor._prepare_merged_action] Executing sudo ['yum install -y -d1 --color=never conntrack-tools iptables-nft openssl curl policycoreutils-python-utils kmod; PACKAGES=$(rpm -q conntrack-tools iptables-nft openssl curl policycoreutils-python-utils kmod); if [ $? != 0 ]; then echo "$PACKAGES" | grep \'is not installed\'; echo "Failed to check version for some packages. Make sure packages are not already installed with higher versions. Also, make sure user-defined packages have rpm-compatible names. "; exit 1; fi '] on host 10.102.0.126 with options: {'warn': False, 'hide': True, 'pty': True, 'env': None, 'timeout': 2700} 10.102.0.126: code=0 === stdout === Last metadata expiration check: 0:00:02 ago on Thu 26 Mar 2026 11:05:04 AM UTC. Package openssl-1:3.5.1-3.el10.x86_64 is already installed. Package curl-8.12.1-2.el10.x86_64 is already installed. Package policycoreutils-python-utils-3.9-1.el10.noarch is already installed. Package kmod-31-12.el10.x86_64 is already installed. Dependencies resolved. ======================================================================================================================================================================================================== Package Architecture Version Repository Size ======================================================================================================================================================================================================== Installing: conntrack-tools x86_64 1.4.8-3.el10 updatesrocky10-yumsrv 235 k iptables-nft x86_64 1.8.11-12.el10_1 updatesrocky10-yumsrv 189 k kernel-core x86_64 6.12.0-124.40.1.el10_1 rocky10-yumsrv 18 M Upgrading: curl x86_64 8.12.1-2.el10_1.2 rocky10-yumsrv 217 k iptables-libs x86_64 1.8.11-12.el10_1 rocky10-yumsrv 408 k libcurl x86_64 8.12.1-2.el10_1.2 rocky10-yumsrv 368 k openssl x86_64 1:3.5.1-7.el10_1 rocky10-yumsrv 1.2 M openssl-fips-provider x86_64 1:3.5.1-7.el10_1 rocky10-yumsrv 812 k openssl-libs x86_64 1:3.5.1-7.el10_1 rocky10-yumsrv 2.2 M Installing dependencies: kernel-modules x86_64 6.12.0-124.40.1.el10_1 rocky10-yumsrv 41 M kernel-modules-core x86_64 6.12.0-124.40.1.el10_1 rocky10-yumsrv 30 M kernel-modules-extra x86_64 6.12.0-124.40.1.el10_1 rocky10-yumsrv 2.8 M libnetfilter_cthelper x86_64 1.0.1-1.el10 updatesrocky10-yumsrv 23 k libnetfilter_cttimeout x86_64 1.0.0-27.el10 updatesrocky10-yumsrv 23 k libnetfilter_queue x86_64 1.0.5-9.el10 updatesrocky10-yumsrv 28 k libnftnl x86_64 1.2.8-4.el10 rocky10-yumsrv 84 k Transaction Summary ======================================================================================================================================================================================================== Install 10 Packages Upgrade 6 Packages Total download size: 97 M Downloading Packages: -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- Total 46 MB/s | 97 MB 00:02 Running transaction check Transaction check succeeded. Running transaction test Transaction test succeeded. Running transaction kdump: For kernel=/boot/vmlinuz-6.12.0-124.40.1.el10_1.x86_64, crashkernel=2G-64G:256M,64G-:512M now. Please reboot the system for the change to take effect. Note if you don't want kdump-utils to manage the crashkernel kernel parameter, please set auto_reset_crashkernel=no in /etc/kdump.conf. Upgraded: curl-8.12.1-2.el10_1.2.x86_64 iptables-libs-1.8.11-12.el10_1.x86_64 libcurl-8.12.1-2.el10_1.2.x86_64 openssl-1:3.5.1-7.el10_1.x86_64 openssl-fips-provider-1:3.5.1-7.el10_1.x86_64 openssl-libs-1:3.5.1-7.el10_1.x86_64 Installed: conntrack-tools-1.4.8-3.el10.x86_64 iptables-nft-1.8.11-12.el10_1.x86_64 kernel-core-6.12.0-124.40.1.el10_1.x86_64 kernel-modules-6.12.0-124.40.1.el10_1.x86_64 kernel-modules-core-6.12.0-124.40.1.el10_1.x86_64 kernel-modules-extra-6.12.0-124.40.1.el10_1.x86_64 libnetfilter_cthelper-1.0.1-1.el10.x86_64 libnetfilter_cttimeout-1.0.0-27.el10.x86_64 libnetfilter_queue-1.0.5-9.el10.x86_64 libnftnl-1.2.8-4.el10.x86_64 Complete!

And with our proposed changes we are detecting such kernel upgrade and triggering reboot during kubemarine.procedures.install.manage_custom_packages as below-

[install.manage_custom_packages] Skipped - no packages configuration defined in config file [executor.queue] Performing sudo ('uname -r',) on nodes ['10.102.0.126'] with options: {'warn': False, 'hide': True, 'pty': False, 'env': None, 'timeout': None} [executor._prepare_merged_action] Executing sudo ['uname -r'] on host 10.102.0.126 with options: {'warn': False, 'hide': True, 'pty': False, 'env': None, 'timeout': 2700} [executor.queue] Performing sudo ('rpm -q kernel-core',) on nodes ['10.102.0.126'] with options: {'warn': True, 'hide': True, 'pty': False, 'env': None, 'timeout': None} [executor._prepare_merged_action] Executing sudo ['rpm -q kernel-core'] on host 10.102.0.126 with options: {'warn': True, 'hide': True, 'pty': False, 'env': None, 'timeout': 2700} [system.detect_kernel_upgrade] 10.102.0.126: Kernel upgrade detected (running kernel: 6.12.0-124.8.1.el10_1.x86_64, latest installed: 6.12.0-124.40.1.el10_1.x86_64) [install.system_prepare_package_manager_manage_packages] Rebooting node(s) 10.102.0.126 to apply updated kernel before proceeding with further system configuration. [__init__.is_cluster_installed] Searching for already installed cluster... [executor.queue] Performing sudo ('kubectl cluster-info',) on nodes ['10.102.0.126'] with options: {'warn': True, 'hide': True, 'pty': True, 'env': None, 'timeout': 15} [executor._prepare_merged_action] Executing sudo ['kubectl cluster-info'] on host 10.102.0.126 with options: {'warn': True, 'hide': True, 'pty': True, 'env': None, 'timeout': 15} [__init__.is_cluster_installed] Failed to detect any Kubernetes cluster [executor.queue] Performing sudo ('last reboot',) on nodes ['10.102.0.126'] with options: {'warn': False, 'hide': True, 'pty': False, 'env': None, 'timeout': None} [executor._prepare_merged_action] Executing sudo ['last reboot'] on host 10.102.0.126 with options: {'warn': False, 'hide': True, 'pty': False, 'env': None, 'timeout': 2700} [executor.queue] Performing sudo ('systemctl stop sshd || sudo systemctl stop ssh ; sudo reboot 2>/dev/null >/dev/null',) on nodes ['10.102.0.126'] with options: {'warn': True, 'hide': True, 'pty': False, 'env': None, 'timeout': None} [executor._prepare_merged_action] Executing sudo ['systemctl stop sshd || sudo systemctl stop ssh ; sudo reboot 2>/dev/null >/dev/null'] on host 10.102.0.126 with options: {'warn': True, 'hide': True, 'pty': False, 'env': None, 'timeout': 2700} [system.perform_group_reboot] Waiting for boot up... [system.perform_group_reboot] Initial boot history: [system.perform_group_reboot] 10.102.0.126: code=0 [system.perform_group_reboot] === stdout === [system.perform_group_reboot] reboot system boot 6.12.0-124.8.1.e Thu Mar 26 11:03 still running [system.perform_group_reboot] [system.perform_group_reboot] wtmp begins Thu Mar 26 11:03:58 2026 [system.perform_group_reboot] [executor._wait_for_boot_with_executor] Trying to connect to nodes, timeout is 600 seconds... [executor._disconnect] Disconnected session with 10.102.0.126 .... [executor._wait_for_boot_with_executor] Attempting to connect to nodes... [executor._prepare_merged_action] Executing run ["sudo -S -p '[sudo] password: ' last reboot"] on host 10.102.0.126 with options: {'hide': True, 'pty': True, 'watchers': [<kubemarine.core.executor.RawExecutor._do_nopasswd.<locals>.NoPasswdResponder object at 0x78b8c14903b0>], 'timeout': 2700} [executor._wait_for_boot_with_executor] All nodes are online now

pranavcracker added 3 commits March 18, 2026 17:42

support rocky linux 10.1

ac55a40

early reboot in case of kernel upgrade

fb761a3

fix ci

c7e7174

pranavcracker marked this pull request as ready for review March 26, 2026 11:55

pranavcracker requested review from nikhil1697 and theboringstuff March 26, 2026 11:55

pranavcracker assigned DmitriiRabenok Mar 26, 2026

theboringstuff reviewed Mar 27, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add support for Rocky Linux 10.1#797

Add support for Rocky Linux 10.1#797
pranavcracker wants to merge 3 commits intomainfrom
feat/rocky_10.1

pranavcracker commented Mar 18, 2026 •

edited

Loading

Uh oh!

theboringstuff Mar 27, 2026

Uh oh!

pranavcracker Mar 27, 2026 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

pranavcracker commented Mar 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Background

Solution

Changes Introduced

1. Rocky Linux 10.1 Support

2. Kernel Upgrade Detection

3. Early Reboot Mechanism

4. Avoid Duplicate Reboots

How to apply

Test Cases

Checklist

Unit tests

Uh oh!

theboringstuff Mar 27, 2026

Choose a reason for hiding this comment

Uh oh!

pranavcracker Mar 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

pranavcracker commented Mar 18, 2026 •

edited

Loading

pranavcracker Mar 27, 2026 •

edited

Loading