Skip to main content

Enable Kdump since first boot

This guide outlines the steps to automatically enable and configure kdump on a SUSE Linux Enterprise Server (SLES) system as part of an installation using Agama. This ensures that kdump is active and ready to capture kernel crash dumps already at the first system boot, providing crucial diagnostic information in case of system failures.

Kdump configuration​

Early Kdump configuration can be done integrating a post-installation script at the Agama configuration. That allows to execute some commands within the chrooted environment of the installed system. Let's see an example using a Jsonnet profile (more readable than plain JSON).

{
  product: {
    id: 'SLES',
    registrationCode: 'YOUR-SLES-REGISTRATION-CODE',
  },
  user: {
    fullName: 'Monsignore Tux',
    password: 'linux',
    userName: 'tux',
  },
  scripts: {
    post: [
      {
        name: 'enable-kdump',
        chroot: true,
        content: |||
          #!/usr/bin/bash
          zypper -n in kdump
          systemctl enable kdump-commandline.service
          systemctl enable kdump.service
          kdumptool commandline -u
        |||,
      },
    ],
  },
}

Let's break down the commands within the content of the post script:

  • #!/usr/bin/bash: Shebang line, ensuring the script is executed with bash.
  • zypper -n in kdump: This command uses zypper (SUSE's package manager) to install the kdump package. The -n flag stands for "non-interactive" and assumes "yes" to any prompts, which is essential for unattended installations.
  • systemctl enable kdump-commandline.service: This enables an auxiliary service that is crucial for setting up Kdump parameters from the kernel command line during boot.
  • systemctl enable kdump.service: This enables the main Kdump service, which manages the Kdump daemon responsible for capturing and saving crash dumps.
  • kdumptool commandline -u: This updates the bootloader configuration to ensure it contains the proper kernel command-line arguments with recommended values.

Integrate the configuration into your installation​

If you are performing an unattended installation, simply merge the scripts section from the provided snippet into your existing Agama configuration profile. The following example shows how to do that using a JSON profile.

{
  "product": {
    "id": "SLES",
    "registrationCode": "YOUR-SLES-REGISTRATION-CODE"
  },
  "user": {
    "fullName": "Your Name",
    "password": "your_password",
    "userName": "your_username"
  },
  "scripts": {
    "post": [
      {
        "name": "enable-kdump",
        "chroot": true,
        "content": "#!/usr/bin/bash\nzypper -n in kdump\nsystemctl enable kdump-commandline.service\nsystemctl enable kdump.service\nkdumptool commandline -u"
      }
    ]
  }
}

If you are using Agama interactively, the corresponding section of the starter's guide explains how to use a partial profile to configure settings that are not available at the web user interface. That can be used to setup our kdump post script. Simply save the scripts section into a separate file (e.g., kdump_config.jsonnet or kdump_config.json if you prefer plain JSON).

{
  scripts: {
    post: [
      {
        name: 'enable-kdump',
        chroot: true,
        content: |||
          #!/usr/bin/bash
          zypper -n in kdump
          systemctl enable kdump-commandline.service
          systemctl enable kdump.service
          kdumptool commandline -u
        |||,
      },
    ],
  },
}

Then, during your interactive Agama session, load that configuration as usual.

agama config load < kdump_config.jsonnet

Finally proceed normally with the rest of the interactive installation.

Verification after installation​

After the SLES installation is complete and the system has rebooted for the first time, you can verify that kdump is enabled and configured correctly:

  1. Check Kdump Service Status:

    sudo systemctl status kdump
    sudo systemctl status kdump-commandline.service

    Both services should be active (exited) or active (running).

  2. Verify Kdump Kernel Parameter: Check the kernel command line to ensure crashkernel= allocation is present.

    cat /proc/cmdline

    You should see an entry like crashkernel=<size>M@<offset>.

  3. Test Kdump (Optional, on a non-production system): To confirm Kdump functionality, you can intentionally trigger a kernel crash. Do not do this on a production system.

    echo c | sudo tee /proc/sysrq-trigger

    After the system reboots, check the /var/crash directory for a new crash dump.