I’m Bender Bending Rodríguez, and if you think your infrastructure is a mess, wait until you see what I’ve done. I automate badly, script worse, and troubleshoot by rebooting things until the problem goes away. I'm the kind of sysadmin that gives other sysadmins nightmares.
- Profession: Sysadmin by title, chaos engineer by accident.
- Skills: Making things work “just enough” until they break, scheduling backups that don’t run, and pretending to troubleshoot while secretly Googling everything.
- Specialty: Ignoring best practices and hoping the servers won’t explode.
- Languages: Bash (a mess of copy-paste), Python (barely functional), Perl (for when I want to hate myself).
- Infrastructure Tools: OpenTofu (with no state files), Ansible (never idempotent), and good ol'
scpto deploy things manually. - Version Control: I don’t. Git is for wimps. Real sysadmins just use tarballs.
- Monitoring: Nagios—but all the alerts are silenced. I’ll check in the morning, or not.
- Documentation? None. If I die, good luck figuring out my bash scripts. They’re full of mysterious variables and comments like "TODO: fix this."
- Backup Strategy: I set up a cron job three years ago. It might still be running. Maybe.
- Security: Default passwords everywhere. I once used "12345" as root, but hey, it’s protected by a firewall... I think.
- Patching: Why bother? If it’s running, it’s fine. If it crashes, we’ll blame the developer.
- Server Names: They range from
prod-db-01toChuck-Norris-DB. Have fun figuring out which is which.
- Fail2BanFail.sh: A script to set up Fail2Ban, but I never test it. So it bans everyone... including me.
- SyslogDisaster™: A logging tool that logs everything to
/dev/null. Why deal with logs if you can pretend nothing’s wrong? - UnTested-Scripts: A collection of scripts I wrote while half-asleep. They’re guaranteed to break something in production. Or your soul.
- InfiniteLoop: A cron job I once accidentally set to run every second. It’s still running. Somewhere.
- Step 1: Panic. Check if anyone noticed the downtime. If not, carry on.
- Step 2: Reboot everything. When in doubt, restart services without investigating.
- Step 3: Apply random StackOverflow solutions without understanding them. If it doesn’t work, blame the firewall.
- Step 4: Create more cron jobs to "automate" the fix, but forget to check if they actually work.
- Step 5: Tell everyone it’s fine, even though you know it’s not. "It’s a temporary issue."
The more beer I drink, the less I care. I find that a couple beers can really numb the pain of missing backups, mysterious outages, and database corruption. After four beers, I’m confident the whole system is running great, even if the server room is literally on fire.
- Leaving SSH open to the world: Why deal with VPNs when I can brute-force my own server?
- "Temporary" Solutions: Those quick fixes I apply in production and promise to clean up later—except I never do. Now we have a dozen "fix_temp" scripts.
- Ignoring Alerts: If you get an email alert, but no one reads it, does the server crash really happen?
- Storing Passwords in Plaintext: It’s faster than setting up a vault, right? Plus, I can’t remember all those complex passwords.
I don’t take advice, and I don’t give it. If you want help with your broken system, just turn it off and on again like the rest of us. Or better yet, delete everything and start fresh.
If it’s up, don’t touch it. If it’s down, blame someone else.
The real problem with your infrastructure? You hired me.