SP> What are you using to protect your computer and bbs
SP> like peerblock firewall, pfsense- with like a check point hardware or
SP> any other hardware? I receive a lot of hits from Russia, Korean
SP> Republic, China etc. So I thought I would see what everyone is
Well, for what it's worth, before I put up my board I was interested in what
exactly these were, so using Netcat and a shell script, I made a kind of
honeypot which prints a login and password prompt, logs those, then prints a
fake shell prompt ($ or # depending on the attempted login).
Nearly all hits to telnet ports are bots/worms spraying-and-praying across the
net, looking for -- so far as I can tell -- cheapo security cameras and other
IoT devices with known default logins and passwords. (I could determine this
by watching what login/password combinations were being tried, then searching
for devices with known defaults of these combinations)
Most are webcams - for some reason - with brand names they don't sell in my
country - as to your comment, most are from places like China and Russia.
Once they are "logged in," nearly all of them attempt to run busybox with a
payload. Some attempt to wget the payload from an external site although for
some reason those have mostly faded away. The busybox command line assumes the
payload is already baked into busybox (i.e. the device already has a
compromised busybox executable).
The scripts are rather dumb; they don't check for result text or error text
from the commands they run.
The larger point here is that unless you're running a system with common
default logins and passwords, these present no threat to your system. They are
nuisances.
Moving your system off of the default ports completely stops them, since these
scripts are looking for low-hanging fruit and targets of opportunity. This
isn't really security-through-obscurity so much as it is moving out of the way
of an indiscriminately fired machine gun.
fail2ban and similar techniques are fine as far as they go but there are so
many of these coming from so many different IP addresses, it's whack-a-mole.
Maybe since it is automated, no big deal.
There's no real threat here. Not that better security is a bad thing; have at
it, but I figured I'd post this just to provide some additional information.
Of the ports I watch (basically everything in /etc/services), these are the
most common hits (note the most hammered port -- hence the issue SysOps have to
put up with):
| Port | Hits | Description
23 37940 telnet
22 27589 ssh - SSH Remote Login Protocol
443 20170 https - http protocol over TLS/SSL
80 18976 http www - WorldWideWeb HTTP
123 15946 ntp - Network Time Protocol
389 5430 ldap - Lightweight Directory Access Protocol
111 2711 sunrpc portmapper - RPC 4.0 portmapper
21 2465 ftp
67 2448 bootps
68 2291 bootpc
1194 1687 openvpn
873 1132 rsync
None of the ports you see in this list are open/provide services on the servers
I monitor, so no one should be legitimately hitting them.
The other traffic you see are from research/scanning IPs - shodan.io is one,
which are people mapping the net or searching for vulnerabilities - generally
good guys (like Arbor Observatory).
Anyway slightly off-topic to your question but I hope there's something
interesting in here of interest to someone.
--- Mystic BBS v1.12 A47 2021/09/24 (Linux/64)
* Origin: Shipwrecks & Shibboleths [San Francisco, CA - USA] (1:218/860)
|
