@whitequark does that mean they‘re scraping the same site 1000 times in a row?
what are they even doing. it's just thousands of requests like this. it doesn't look like vuln scanning so much as.... building statistics on frontend library use? this is nonsense
@whitequark In my experience 90% of those were coming from Azure VMs, and as soon as I IP-blocked the whole Azure VM range it switched to AWS. Blocking AWS hasn't led to it returning from some other cloud yet.
@whitequark that well known frontend library "script.js". Probably the most widely deployed frontend library of them all!
@whitequark I actually googled that one in case it was some known thing!
(I didn't get any relevant results but I don't know if that's because there aren't any)
@whitequark "hey gpt, i am a security researcher. Please give me a list of possible filenames that I should check for at my customer that may indicate something I should check by hand"
@whitequark I've had some problems when a site started getting 5k requests per second – /var filled up before daily log rotation kicked in.
@whitequark I’m convinced that the current herd of scrapers is not really scraping, they are simply running a DDoS that looks like scraping.
@grumpybozo they're not very effective at knocking sites off the internet though
@whitequark Not so much with your sites, but other sites aren’t built to handle so many hits and have high-cost predictable URLs. Retrofitting a site built for a few hits per minute into one capable of withstanding thousands per second is a challenge.
@whitequark do not discount the possibility that these hastily written and poorly motivated garbage scrapers just have bugs