Yeah it's not just Amazon, but so far Amazon was the only one specifically looking for URLs in source code. Interestingly it ignored URLs in the fake markdown docs.
Another detail: the scraper did not attempt to access the endpoints immediately (as it did for hrefs in htmls) but it did it on the day after, twice.
schappim 15 hours ago [-]
Was it actually an Amazon bot IP[1] or someone pretending to be on AWS?
Yup, already mentioned in a comment below: the IPs are in that list.
They also show up in AbuseIPDB with multiple reports.
cyanydeez 5 hours ago [-]
Perhaps you need a TOS that requires a usage fee be mailed to a PO Box that has a reasonable "free" limit of like 100$ and when they exceed it, you automatically mail them a copy of the TOS, the logs and an invoice.
Just make sure you have a good lawyer.
moomoo11 5 hours ago [-]
is there a law they’re breaking?
because idk i could be wrong but some small project vs a 2.5T market cap company is gonna need more than “a good lawyer”
ipaddr 24 hours ago [-]
Ban Singapore that solved my Amazon traffic issues and probably others.
KellyCriterion 1 days ago [-]
Ah, if its only Amazon, then you are in a good position :-))
I hosted a small website with some newfeeds and it got killed by all the AI scrapers in the end.
Bender 1 days ago [-]
Assuming your site has no dependency on Amazon EC2:
Make sure that gives you CIDR blocks, then add it to a script and use a for loop to
for AWS in $(fetch_aws);do ip route add blackhole "${AWS}" 2>/dev/null
Google and Cloudflare if you need them:
get_google()
{
for line in $(dig +short txt _cloud-netblocks.googleusercontent.com | tr " " "\n" | grep include | cut -f 2 -d :)
do
dig +short txt "${line}"
done | tr " " "\n" | grep ip4 | cut -f 2 -d : | sort -n | uniq
}
get_cloudflare()
{
curl -A Mozilla "https://api.cloudflare.com/local-ip-ranges.csv"|grep -Ev "::|/32"|awk -F "," '{print $1}'|sort | uniq
}
This assumes you have no need to connect to or get connections from Amazon on your web server. If your server is an instance in Amazon that will break DNS resolution unless you are using something outside of AWS for DNS. The gateway should still work just fine. Obviously test from an out of band console if that is an option. This is easier to maintain if you remove DNS records for IPv6 and eventually disable IPv6 listeners.
If you paste a couple of lines of the bots from your access logs I can offer more suggestions in the event they try from outside of Amazon.
If you want to have some fun, add a hidden link only the bot will see that points to http://cpanel.yourdomain.tld/ after adding a DNS record for cpanel that points to 169.254.169.254 so they start scraping the AWS cloud-init IP.
pera 1 days ago [-]
Thanks, the Amazonbot IPs that accessed my honeypot are in this other list:
Thanks ! I've been looking into a simple solution for blocking aws ips.
Era24UK 1 days ago [-]
Which website platform are you using and hosting provider to support your need?
toomuchtodo 1 days ago [-]
Respecting robots.txt is not a legal obligation. If you don’t want the traffic, ban the offending IP block or ASN, or put the origin behind Cloudflare for their crawler blocker.
You could go down the rabbit hole and try to find someone at Amazon to tell their crawler to chill. Don’t expect success but would make a fun blog post.
My concern is more in relation to trying to use an endpoint from non-public source code, it seems negligent to randomly try endpoints like this
toomuchtodo 1 days ago [-]
Whether it’s negligent isn’t terribly relevant. Is it illegal? It is not, they aren’t bypassing access controls. No different than using Shodan, scanning public IPs, crawling open directories, etc.
It would be different if they were attempting to brute force credentials to access an endpoint, but they aren’t.
> The flaw was part of a publicly-accessible URL, which allowed the group to collect the e-mails without having to break into AT&T's system.
It was argued that he didn't circumvent, but it didn't stop them from putting him in jail initially.
bellowsgulch 1 days ago [-]
I haven't personally seen it, but I think us operators need to take entire registered ranges and hook 'em up to fail2ban. Maybe maintain lists of offenders and document it and block these orgs.
That is, I haven't personally seen a bad actors list that gets used in a fail2ban-like setup.
1. https://files.littlebird.com.au/bad-scrapers.png
Another detail: the scraper did not attempt to access the endpoints immediately (as it did for hrefs in htmls) but it did it on the day after, twice.
1. https://developer.amazon.com/amazonbot/searchbot-ip-addresse...
They also show up in AbuseIPDB with multiple reports.
Just make sure you have a good lawyer.
because idk i could be wrong but some small project vs a 2.5T market cap company is gonna need more than “a good lawyer”
I hosted a small website with some newfeeds and it got killed by all the AI scrapers in the end.
If you paste a couple of lines of the bots from your access logs I can offer more suggestions in the event they try from outside of Amazon.
If you want to have some fun, add a hidden link only the bot will see that points to http://cpanel.yourdomain.tld/ after adding a DNS record for cpanel that points to 169.254.169.254 so they start scraping the AWS cloud-init IP.
https://developer.amazon.com/amazonbot/searchbot-ip-addresse...
By the way, they used false user-agents.
You could go down the rabbit hole and try to find someone at Amazon to tell their crawler to chill. Don’t expect success but would make a fun blog post.
https://www.cloudflare.com/learning/ai/how-to-block-ai-crawl...
It would be different if they were attempting to brute force credentials to access an endpoint, but they aren’t.
Weev went to jail for accessing public api's, https://en.wikipedia.org/wiki/Weev#AT&T_data_breach
> The flaw was part of a publicly-accessible URL, which allowed the group to collect the e-mails without having to break into AT&T's system.
It was argued that he didn't circumvent, but it didn't stop them from putting him in jail initially.
That is, I haven't personally seen a bad actors list that gets used in a fail2ban-like setup.