So i have a domain that I have been using solely for homelab and VPS services (domain.example).
I have my A and AAAA record for my VPS proxying through cloudflare (proxy.domain.example) and a DNS A record pointing towards my homelab for my home Wireguard (wg.domain.example) with no other records pointing home or anywhere. I have a couple of services at home with certificates for example (proxmox.domain.example, nas.domain.example, router.domain.example) that are using cloudflares API token but they do not have records listed at cloudflare
Now my issue is I specifically setup a Cloudflare WAF to block every continent/country except my own and this is now showing in the events that a crawler is attempting to access router.domain.example, nas.domain.example, homeassistant.domain.example. Do I have any reason to be concerned and also how would this web crawler only be searching for my home lab domains. None of these services are public facing.
Probably not. It’s most likely automated scanning and the subdomains seem common enough to be included in wordlists. Another possibility is that the subdomains have leaked somehow, do you use LetsEncrypt? If so, the existence of your subdomains is public knowledge and can easily be picked up by bots.
If anyone is interested in mitigation, the only way around this AFAIK is to start with a brand new domain, only use wildcard certs (with DNS validation), and don’t bundle multiple renewals into a single cert.
Also, don’t enter your domain or related IP address into dns reverse engineering tools (like dnsdumpster), and check certificate transparency logs (https://crt.sh) to see what information related to your cert renewals has been published.
This won’t stop automated bots from scanning your ip for domains, but should significantly reduce the amount of bots that discover them
I think it is generally okay to bundle the root domain certificate and the wildcard for its subdomains into a single renewal.
So for example:
example.com *.example.com
Yepp sorry - what I meant was bundling multiple different root domains, e.g.
example.com
&example1234567.org
in the same cert.I currently do as you mentioned above, renewing with just one root bundled with its accompanying subdomain wildcard.
It’s not just let’s encrypt - the common names of any SSL cert issued by a public CA have to be recorded in a public certificate transparency log. You can use tools like https://crt.sh to search the logs
Ahhhh thank you. Yes I use LetsEncrypt for all the homelab services which explains it then.
Its one reason i use DNS challenge wildcard domains.
I know security through obscurity is not security, and that a leaked wildcard cert is more damaging… However the likelihood of a leaked cert is slim, the convenience is huge, the attack window isn’t huge (well, 90 days) and less published information about internals feels more secure.maybe you issued one certificate with multiple domains, mixing internet facing ones with purely internal. It is very easy to discover hidden subdomains inspecting the certificate you get from a public service
This is my thought as well.
Those services are running on some ports and someone was able to see that there are services running on those ports. Now they (or more likely, their script) is trying to find out what those services/versions are to see if there are exploits.
So to OPs question should they be worried? No. This is par for the course today. But is a great example of why you need to be vigilant in updating your services and platforms, use strong passwords, MFA, etc.
Here’s good piece of guidance for any and all who are managing a domain/network.
The lower on the pyramid of pain you can make it a pain in the ass for a would-be intruder, the sooner they’ll give up. In OPs example, they are moving from ‘Domain names’ to ‘network/host artifacts’ if they fail to get enough info to keep digging down, they’ll likely stay there and persist for awhile and then give up if they don’t find a crack.
If I had to guess after managing enterprise WAF across hundreds of domains…
It’s either a crowler or vulnerability scanner, and may be scanning by IP address. I don’t think you configured anything wrong.
You may want to add some form of captcha or user agent based filter to get rid of it. Good news is that it’s not necessarily something to worry about.
I’d avoid IP based blocking. It’s only temporarily effective.
As long as you set your home server properly, with allow rules, firewalls and stuff, you’ll be fine
But don’t be sloppy, take care of your server
Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:
Fewer Letters More Letters CA (SSL) Certificate Authority DNS Domain Name Service/System IP Internet Protocol SSL Secure Sockets Layer, for transparent encryption
[Thread #613 for this sub, first seen 19th Mar 2024, 10:35] [FAQ] [Full list] [Contact] [Source code]
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator
deleted by creator