What is auto_reset_crashkernel in /etc/kdump.conf and how does it work?

Solution Verified - Updated

Environment

  • Red Hat Enterprise Linux 9
  • kexec-tools-2.0.23-9_1.el9_0 and newer version

Issue

  • RHEL9 introduces auto_reset_crashkernel option in /etc/kdump.conf.
  • How does the parameter work?

Resolution

2.0.25-13.el9_2.1 and older version

  • auto_reset_crashkernel=yes could update the default crashkernel value when updating the kexec-tools package
    in the following two cases;
    1. crashkernel=auto is found in the kernel cmdline
    2. the crashkernel was previously set by kexec-tools i.e. the kernel is using the old default crashkernel value.
  • If the system sets a specific crashkernel value (i.e. crashkernel=256M), the option does not update crashkernel value.

2.0.26-8.el9 and newer version

  • auto_reset_crashkernel yes could update the default crashkernel value when the kexec-tools package
  • Unlike the above old versions, the user-specific crashkernel value will also be overwritten to 1G-4G:192M,4G-64G:256M,64G-:512M. To remain the user-specific value, set auto_reset_crashkernel no or re-apply the old user-specific crashkernel value by the following procedure after updating kexec-tools.

Root Cause

  • Because crashkernel=auto is deprecated in RHEL9, kexec-tools needs to update the old default value to the new default value 1G-4G:192M,4G-64G:256M,64G-:512M. For more details about the crashkernel=auto, please check the following KCS.
  • Since the following commit is applied at kexec-tools-2.0.26-8.el9, crashkernel value will be set a new default value all time when updating kexec-tools if auto_reset_crashkernel yes sets.
commit cb850aec2623b66d7361636e806ddf33e9de0eac
Author: Pingfan Liu <piliu@redhat.com>
Date:   Thu Jun 15 10:11:53 2023 +0800

    Simplify the management of the kernel parameter crashkernel

    Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2160676
    Upstream: Fedora rawhide
    Conflict: applied manually due to slight difference in context

    commit 5b31b099ae9b40a8f832b07e8364d7b08025fdd6
    Author: Coiby Xu <coxu@redhat.com>
    Date:   Wed Apr 26 04:48:25 2023 +0800

        Simplify the management of the kernel parameter crashkernel

        Currently, kexec-tools only updates the crashkernel to a new default
        value only when both two conditions are met,
         - auto_reset_crashkernel=yes in kdump.conf
         - existing kernels or current running kernel should use the old default
           value.

        To address seen corner cases, the logic to tell if the second condition
        is met becomes quite complex. Instead of making the logic more complex
        to support aarch64-64k, this patch drops the second condition to
        simplify the management of the crashkernel kernel parameter.

        Another change brought by this simplification is kexec-tools will also
        set up the kernel crashkernel parameter for a fresh install (previously
        it's limited to osbuild).

        Note
        1. This patch also stop trying to update /etc/default/grub because
           a) it only affects the static file /boot/grub2/grub.cfg
           b) grubby is recommended to change the kernel command-line parameters
              for both Fedora [1] and RHEL9 [2][3]
           c) For the cases of aarch64 and POWER, different kernels could have
              different default crashkernel value.

        2. Starting with Fedora 37,  posttrans rpm scriplet distinguish between
           package install and upgrade.

        [1] https://fedoraproject.org/wiki/GRUB_2
        [2] https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/9/html/managing_monitoring_and_updating_the_kernel/configuring-kernel-command-line-parameters_managing-monitoring-and-updating-the-kernel#changing-kernel-command-line-parameters-for-all-boot-entries_configuring-kernel-command-line-parameters
        [3] https://access.redhat.com/solutions/1136173

        Signed-off-by: Coiby Xu <coxu@redhat.com>
        Reviewed-by: Philipp Rudo <prudo@redhat.com>

    Signed-off-by: Pingfan Liu <piliu@redhat.com>

Diagnostic Steps

  • The old default value is updated.
# cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt3)/vmlinuz-5.14.0-70.13.1.el9_0.x86_64 
(snip) net.ifnames=0 crashkernel=auto

# cat /etc/kdump.conf |grep reset|grep -v ^#
auto_reset_crashkernel yes

# rpm -qa kexec-tools
kexec-tools-2.0.23-9_1.el9_0.x86_64

# dnf install kexec-tools -y
# reboot

# cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt3)/vmlinuz-5.14.0-70.13.1.el9_0.x86_64 
(snip) net.ifnames=0 crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M

# rpm -qa kexec-tools
kexec-tools-2.0.25-13.el9_2.1.x86_64
  • The crashkernel set a specific value is not changed when updating kexec-tools to 2.0.25-13.el9_2.1.
# cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt3)/vmlinuz-5.14.0-70.13.1.el9_0.x86_64 
(snip) net.ifnames=0 crashkernel=256M

# cat /etc/kdump.conf |grep reset|grep -v ^#
auto_reset_crashkernel yes

# rpm -qa kexec-tools
kexec-tools-2.0.23-9_1.el9_0.x86_64

# dnf install kexec-tools -y
# reboot

# cat /proc/cmdline
BOOT_IMAGE=(hd0,gpt3)/vmlinuz-5.14.0-70.13.1.el9_0.x86_64 
(snip) net.ifnames=0 crashkernel=256M

# rpm -qa kexec-tools
kexec-tools-2.0.25-13.el9_2.1.x86_64
  • But if the auto_reset_crashkernel yes, it will be overwritten when updating it to 2.0.26-8.el9 or later
# cat /proc/cmdline
(snip) crashkernel=256M

# dnf install kexec-tools-2.0.26-8.el9 -y
# reboot

# cat /proc/cmdline 
(snip) net.ifnames=0 crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M
SBR
Product(s)
Components
Category
Tags

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.