RHEL Lightspeed

AI-Driven Server Management

CIS126RH | RHEL System Administration
Mesa Community College

Learning Objectives

What is RHEL Lightspeed?

How Lightspeed Works

Accessing Lightspeed

# Prerequisite: System must be registered with Red Hat
[root@server ~]# subscription-manager status
System Status: Current

# Install web console (Cockpit) if not present
[root@server ~]# dnf install cockpit
[root@server ~]# systemctl enable --now cockpit.socket

# Access web console at https://your-server:9090
# Look for the Lightspeed icon in the interface

Web Console Interface

Command Line Access

# Ask Lightspeed a question from the command line
[root@server ~]# lightspeed "How do I restart the Apache web server?"

To restart the Apache web server on RHEL, use:

  systemctl restart httpd

To check the status after restarting:

  systemctl status httpd

If you want Apache to start automatically at boot:

  systemctl enable httpd

CLI Examples — find and explain

# Quick command lookup
[root@server ~]# lightspeed "command to find files larger than 100MB"

find / -type f -size +100M -exec ls -lh {} \;

This searches from root (/) for files (-type f) larger than 100MB
(-size +100M) and displays details for each match.

# Ask Lightspeed to explain a command you don't understand
[root@server ~]# lightspeed "explain: tar -czvf backup.tar.gz /home"

This command creates a compressed archive:
  tar     - tape archive utility
  -c      - create new archive
  -z      - compress with gzip
  -v      - verbose (show files being archived)
  -f      - specify filename (backup.tar.gz)
  /home   - directory to archive

Task: User Management

Task: Service Management

Task: Firewall Configuration

Task: Storage Management

[root@server ~]# lightspeed "extend the root logical volume by 5GB"

To extend the root LV by 5GB, follow these steps:

1. Check available space in the volume group:
   vgs

2. Extend the logical volume:
   lvextend -L +5G /dev/mapper/rhel-root

3. Resize the filesystem (for XFS):
   xfs_growfs /

   For ext4 filesystems use instead:
   resize2fs /dev/mapper/rhel-root

4. Verify the new size:
   df -h /

Storage: Follow-up Questions

[root@server ~]# lightspeed "my volume group has no free space, options?"

Options when VG has no free space:

1. Add a new disk/partition as physical volume:
   pvcreate /dev/sdb1
   vgextend rhel /dev/sdb1

2. Reduce another logical volume (if possible):
   lvreduce -L -5G /dev/mapper/rhel-home
   (Warning: resize filesystem FIRST before reducing LV!)

3. Add a virtual disk (if VM) and extend

4. Use thin provisioning for future flexibility

Task: Security & SELinux

Learning with Lightspeed

[root@server ~]# lightspeed "explain the difference between systemctl start and enable"

These commands serve different purposes:

systemctl start <service>
  - Starts the service immediately
  - Does NOT persist across reboots
  - Use when you need the service running now

systemctl enable <service>
  - Creates symlinks so service starts at boot
  - Does NOT start the service immediately
  - Use to ensure service runs after reboot

Common pattern - do both:
  systemctl enable --now <service>

The --now flag combines both: enables AND starts immediately.

Deep Concepts with Lightspeed

[root@server ~]# lightspeed "what are cgroups and why do they matter?"

Cgroups (control groups) are a Linux kernel feature that:

- Limit resources (CPU, memory, I/O) for process groups
- Isolate processes from each other
- Account for resource usage
- Enable containerization (Docker, Podman use cgroups)

In RHEL, systemd uses cgroups v2 to manage service resources.
You can limit a service's memory with:
  systemctl set-property httpd MemoryMax=512M

Writing Effective Prompts

Verifying Suggestions

# Before running any suggested command:

# 1. Ask Lightspeed to explain if you're unclear about what it does
[root@server ~]# lightspeed "explain: rm -rf /var/log/old/*"

# 2. Check if paths/service names match YOUR system
#    - Paths may differ from the suggestion
#    - Service names can vary by version
#    - Your system config may differ from defaults

# 3. For destructive operations, verify paths before executing
[root@server ~]# ls /var/log/old/          # Confirm directory exists and contains expected files
[root@server ~]# rm -rf /var/log/old/*      # Then execute

Limitations and Boundaries

Workflow Integration

Practical Examples

# "How do I..." questions
"How do I schedule a job to run every Sunday at 3am?"
"How do I check which process is using port 8080?"
"How do I mount an NFS share permanently?"

# Troubleshooting questions
"SSH connection times out to server, but ping works fine"
"Disk shows full but du says only 50% used"
"Service starts then stops immediately"

# Explanation requests
"Explain what this cron entry means: */15 * * * *"
"What's the difference between kill and kill -9?"
"Why does rsync use -a flag by default?"

# Best practice questions
"What's the recommended way to update packages safely?"
"How should I structure sudoers for a dev team?"
"Best practices for SSH key management?"

Security Considerations

Best Practices

Key Takeaways

Next: Navigating the Filesystem Hierarchy