10_check_system_health.yml - Usage Guide

Read-Only

Purpose

This playbook reports overall system health for a server:

  • OS (distribution/version), kernel version, uptime
  • CPU core count
  • Load average (1/5/15 min) and 1-minute load normalized per core
  • Run queue (currently running/total process count, from /proc/loadavg)
  • Zombie (defunct) process count
  • Virtualization detection: ansible_facts (role/type) + systemd-detect-virt + dmidecode (manufacturer/product/BIOS version) — tries to identify the hypervisor (Hyper-V, VMware, KVM, and so on)
  • Memory usage (free -h, with automatic MB/GB units)
  • Disk usage (df -hT, real filesystems only; virtual ones such as tmpfs/overlay are excluded), size in MB/GB and percent full
  • Top 10 CPU-consuming processes
  • Top 10 RAM-consuming processes
  • Whether a reboot is required (/var/run/reboot-required)
  • Failed systemd services
  • Count of installable package/security updates

Requirements

  • hosts: all — runs on every node.
  • This playbook uses gather_facts: true (unlike the other playbooks in this repo) — OS/kernel/CPU/virtualization data comes from Ansible’s own fact gathering.
  • dmidecode output needs become: true (sudo); if it is missing or inaccessible, that section quietly shows “could not be accessed” and the playbook does not fail.
  • The first block of the report is machine identity (HOST = inventory name/IP, hostname = the server’s own name); use it to tell which output belongs to which node.

How to run 

ansible-playbook -i inventories/musteri_a/hosts.ini playbooks/10_check_system_health.yml

# Limit to a host or group:
ansible-playbook -i inventories/musteri_a/hosts.ini playbooks/10_check_system_health.yml --limit worker1

Sample output 

TASK [System health report] ****************************************************
ok: [203.0.113.10] => {
    "msg": [
        "################################################################################\n# HOST: 203.0.113.10\n# hostname: node1\n################################################################################",
        "OS: Ubuntu 22.04 (jammy)",
        "Kernel: 5.15.0-186-generic",
        "Uptime: up 2 days, 6 hours, 20 minutes",
        "CPU cores: 16",
        "Load average (1/5/15 min): 0.97 / 0.98 / 1.01 (1-min per core: 0.06)",
        "Run queue (running/total processes): 1/2090",
        "Zombie (defunct) process count: 0",
        "Virtualization: role=guest, type=VirtualPC, systemd-detect-virt=microsoft",
        "Hardware/Hypervisor info:\nManufacturer: Microsoft Corporation\nProduct: Virtual Machine\nBIOS Version: 090007",
        "Memory:\n               total  used  free  shared  buff/cache  available\nMem:  125Gi  5.2Gi  90Gi  121Mi   29Gi        119Gi\nSwap: 0B     0B    0B",
        "Disk usage:\nFilesystem  Type  Size  Used  Avail  Use%  Mounted on\n/dev/mapper/ubuntu--vg-ubuntu--lv  ext4  540G  388G  130G  75%  /",
        "Top 10 CPU-consuming processes:\n...",
        "Top 10 RAM-consuming processes:\n...",
        "Reboot required: YES",
        "Failed systemd services: None",
        "Installable package updates: 32"
    ]
}

Notes

  • systemd-detect-virt + dmidecode together reveal the hypervisor on guest VMs (for example Microsoft Hyper-V, VMware, KVM) and sometimes the BIOS version; the hypervisor’s own version is usually not fully visible from inside the guest — BIOS/product info is only a hint.
  • “Reboot required” and “installable package updates” are Debian/Ubuntu-specific (/var/run/reboot-required, apt list --upgradable); on other distributions those steps quietly return empty/without error.