Linux System Administrator Q&A

How do you get the kernel version / Command to list OS version in Linux?
- To check the kernel version: uname -r
- To list the OS version: cat /etc/os-release or lsb_release -a (if available).
How do you identify if you are working on a physical server or VM?
- Run virt-what (if installed) to detect virtualization.
- Alternatively, check dmesg | grep -i hypervisor or lscpu for signs of a hypervisor (e.g., VMware, KVM).
- You can also use dmidecode -s system-manufacturer or dmidecode -t system to see hardware details—physical servers often show real vendor names (e.g., Dell, HP), while VMs show virtualized names (e.g., VMware, QEMU).
How do you set kernel parameters / Where can I find the file with active parameters?
- Active kernel parameters are stored in /proc/sys/. You can view them with sysctl -a.
- To set a parameter temporarily: sysctl <parameter>=<value> (e.g., sysctl net.ipv4.ip_forward=1).
How do you set kernel parameters permanently?
- Edit /etc/sysctl.conf or a file in /etc/sysctl.d/ and add the parameter (e.g., net.ipv4.ip_forward=1).
- Apply changes with sysctl -p (reloads settings from /etc/sysctl.conf) or sysctl -w <parameter>=<value> for a temporary change.
How to load modules in Linux / Command to list loaded modules?
- Load a module: modprobe <module_name> (e.g., modprobe nfs).
- List loaded modules: lsmod.
How do you check if a filesystem is read-only?
- Run mount | grep <mount_point> and look for ro (read-only) or rw (read-write).
- Alternatively, use cat /proc/mounts to check the mount options.
How do you force a filesystem check during boot or repair it?
- Force a check on next reboot: touch /forcefsck or shutdown -r -F now.
- To repair manually: Boot into single-user mode, then run fsck /dev/<partition> (e.g., fsck /dev/sda1). Use -y to auto-fix issues.
How do you check CPU power performance in a CFM server?
- Assuming "CFM" is a typo or specific context, to check CPU performance generally: Use lscpu for specs, top or htop for usage, or cat /proc/cpuinfo for detailed info.
- For power/performance tuning, check cpufreq-info (if cpufrequtils is installed).
How to check and fix disk space issues in a volume?
- Check space: df -h (human-readable sizes).
- Find large files: du -h /path | sort -rh | head.
- Fix: Delete unnecessary files, expand the volume (e.g., lvextend for LVM), or resize the filesystem (e.g., resize2fs).
Commands to monitor server performance (CPU, memory, disk)?
- top or htop: Real-time CPU and memory usage.
- vmstat 1: CPU, memory, and I/O stats every second.
- iostat: Disk I/O performance (requires sysstat package).
How to check processes and ports?
- Processes: ps aux or top.
- Ports: ss -tuln (listening ports) or netstat -tuln (if installed).
How do you check/set max open files for a specific user?
- Check current limit: ulimit -n (for the current session) or cat /proc/<pid>/limits for a process.
- Set for a user: Edit /etc/security/limits.conf, add <username> soft nofile <value> and <username> hard nofile <value>, then relog.
How to limit the number of files open by a user?
- Edit /etc/security/limits.conf as above (e.g., <username> hard nofile 1024), then apply with sysctl -p or relog the user.
Server not reachable—how to troubleshoot?
- Check if it’s up: ping <server_ip>.
- Try SSH: ssh <user>@<server_ip>—if it fails, check the console (e.g., via ILO/VMware).
- Verify network: ip route (default gateway), ip link (interface status).
Server rebooted but not available after maintenance/OS upgrade?
- Check console for boot messages (stuck at grub, kernel panic?).
- Boot into single-user mode (edit grub with systemd.unit=rescue.target) and review logs (journalctl or /var/log/boot.log).
Filesystem inconsistencies after reboot, goes read-only—how to fix?
- Boot into single-user mode, run fsck /dev/<partition> to repair.
- Remount as read-write: mount -o remount,rw /.
- Check logs (dmesg or journalctl) for root cause.
User can’t create files/folders—what to check?
- Permissions: ls -ld <path> (write access?).
- Inodes: df -i (inodes exhausted?).
- Quota: quota <username> (if quotas are enabled).
Passwordless auth enabled but still prompts for password—how to fix?
- Check ~/.ssh/authorized_keys permissions (600) and directory (700).
- Verify /etc/ssh/sshd_config: PubkeyAuthentication yes, PasswordAuthentication no.
- Restart SSH: systemctl restart sshd.
How to disable a service in Linux?
- Disable at boot: systemctl disable <service>.
- Stop immediately: systemctl stop <service>.
Kernel messages?
- View with dmesg or journalctl -k.
Kernel panic—how to fix?
- Boot into single-user mode (grub: add single or systemd.unit=rescue.target).
- Recreate initrd: mkinitrd or dracut -f (depends on distro), then reboot.
VM is hung, not reachable—how to troubleshoot?
- Check console via hypervisor (e.g., VMware vSphere).
- Restart VM if needed. Review logs (journalctl, dmesg) after recovery.
How to upgrade the kernel version?
- Install new kernel: yum update kernel (RHEL) or apt install linux-image-<version> (Debian).
- Update grub: grub2-mkconfig -o /boot/grub2/grub.cfg (RHEL) or update-grub (Debian), then reboot.
VMware VMs in read-only state due to storage issues—how to fix?
- Check affected VMs: esxcli vm process list or vSphere UI.
- Verify storage: df -h on each VM, fix with fsck if needed, then remount (mount -o remount,rw).
Server in maintenance mode—how to perform filesystem check?
- From console, run fsck /dev/<partition> (unmount first if possible).
- Exit maintenance: reboot or systemctl default.
Backup issue in HANA DB with I/O wait—how to check?
- Check I/O: iostat -x 1 (high %iowait?).
- Review HANA logs and OS logs (journalctl).
- Verify disk space (df -h) and performance in Azure portal (if applicable).
SPC procedure for OS upgrade—how to find backup FS layout in TIC?
- Assuming TIC is a specific system/context, generally: Check /etc/fstab for FS layout.
- Backup layout: lsblk or fdisk -l before upgrade. (Clarify "TIC" for a precise answer.)

Let me know if you need further clarification!