Skip to content

system_administration.md

Latest commit

 

History

History
129 lines (97 loc) · 4.33 KB

system_administration.md

File metadata and controls

129 lines (97 loc) · 4.33 KB
Raw

system administration

process of managing and maintaining computer systems, servers and networks to ensure they run reliably, securely and efficiently

  • responsibilities:
    • installing and configuring operating systems and software
    • managing user accounts and permissions
    • monitoring performance and availability
    • performing backups and recovery
    • applying security patches and updates
    • troubleshooting system issues
    • automating tasks with scripts (like bash)

environments

separate setup where an application runs, used for different stages of development

  • development: where features are developed and tested locally on dev servers

    • used by: developers
    • typical setup: local machines or remote dev servers

  • testing/QA (Quality Assurance): run manual or automated tests on new code before it's promoted further

    • used by: QA testes, developers
    • typical setup: closely mirrors staging, often auto-deployed by CI/CD after commits

  • staging : production-like environment for final testing before going live

    • used by: QA testes, developers, product owners

  • homologation/UAT (User Acceptance Testing): clients/stakeholders test and approve features before production

    • used by: clients, stakeholders, QA
    • very close to production
    • goal: validate that the app behaves correctly in real-world scenarios

  • production: live system accessible to customers

    • used by: end users
    • rules: strict deployment processes, rollback plans, alerting

  • cloning environment: cloning an environment means replicating the full stack — application code, docker setup, environment variables, and database — into a new isolated space, usually for debugging, migration testing or demos

    • normally done with a shell script that copies the project directory, adjusts ports in .env and docker-compose.yml and spins up containers with docker compose up
    • for the database, you can snapshot or dump the data and restore it into a new db container or rds instance, so the clone doesn’t affect the original

deployment strategies

techniques to safely release new versions of an application, minimizing downtime and risk

  • recreate: stop old version, start the new one
    • downtime during switch
    • simple, but not ideal for real time systems
  • rolling update: replace instances gradually, one by one
    • no downtime
    • requires kubernetes or docker swarms or similar
  • blue-green deployment: run two environments, switch traffic when ready
    • blue: current version
    • green: new version
  • canary deployment: release to a small percentage of users, monitor, then expand
    • safer, controlled exposure
  • A/B testing: route users to different app versions to compare
    • analyze which version performs better (e.g. UX, business metrics)

user management

  • useradd: add a new user
  • userdel: delete a user
  • usermod: modify a user
  • passwd: change a user's password

group management

  • groupadd: add a new group
  • groupdel: delete a group
  • groupmod: modify a group

sudo configuration

  • sudo: run a command as another user
  • visudo: edit the sudoers file

scheduling tasks with cron

cron

  • crontab: schedule tasks to run at specific times

mounting and unmounting filesystems

process of attaching a filesystem accessible to a specific point in your system's directory tree

# Basic mount command syntax
mount [device] [mount point]

# Example: mounting a USB drive
mount /dev/sdb1 /mnt/usb

# View mounted filesystems
mount
df -h  # Shows mounted filesystems with human-readable sizes

# Read-only mount
mount -o ro /dev/sdb1 /mnt/usb

# Read-write mount
mount -o rw /dev/sdb1 /mnt/usb

# Unmounting
umount /mnt/usb
  • mount point: empty directories where filesystems are attached
  • common locations:
    • /mnt: temporary mounts
    • /media: removable media
    • /: root filesystem

troubleshooting scenarios

  • laravel app isn't accessible to other services, how to restore accessibility?

    • check if container/aws resource is running
    • check security groups
    • check nginx/apache config
    • check network ACLs (?)
    • check if bound to correct port/IP
    • check env vars

  • how do you check if a container is reachable?

    • curl
    • ping: checks if a host is reachable
    • nc: checks if a port is open
      • scan port 80: nc -zv localhost 80
    • ss
    • netstat