@JeffSchaller That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer Append -force to the command word(s) if you really want to add the pattern you provided.
@tripleee That pattern looks like it's already caught by Blacklisted website in answer and Blacklisted website in body Append -force to the command word(s) if you really want to add the pattern you provided.
@tripleee That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body Append -force to the command word(s) if you really want to add the pattern you provided.
@JeffSchaller That pattern looks like it's already caught by Potentially bad keyword in body and Potentially bad keyword in answer Append -force to the command word(s) if you really want to add the pattern you provided.
[ SmokeDetector | MS ] Pattern-matching website in answer, potentially bad asn for hostname in answer, potentially bad keyword in answer, potentially problematic ns configuration in answer (79): Show taxonomy description on wordpress by arian zagros on stackoverflow.com
@JeffSchaller That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body Append -force to the command word(s) if you really want to add the pattern you provided.
@JeffSchaller I saw that you gave this answer FP feedback. Were you confused by the edit that removed the actual spam link? (The rest of the post is ChatGPT-generated fluff: see my comment and that by Suraj.) cc. @WaiHaLee
@AdrianMole I am below 10k and I rated it fp because there wasn't a spam link in that version on MS. If it's chatgpt then my last understanding --for CHQ-- was fp. I'm aware that SO does not want cgpt posts, but if I'm wrong about the CHQ stance I'd be happy to change my feedback.
The problem was that it had been edited to remove an obvious spam link. (Obvious in the sense that it was "the same" as the username.) I'm not suggestion that you were in error in your feedback.
[ SmokeDetector | MS ] Linked punctuation in body, potentially bad asn for hostname in body, potentially bad ip for hostname in body, potentially bad ns for domain in body, potentially bad keyword in body (22): After changing networks, Wi-Fi does not work by Bts Merch on askubuntu.com
@tripleee That pattern looks like it's already caught by Potentially bad keyword in answer and Potentially bad keyword in body Append -force to the command word(s) if you really want to add the pattern you provided.
@AdrianMole I don't remember that post in particular, but I am pretty sure that I must have done the same thing as Jeff Schaller. I've also cleared my feedback as the opportunity for me to see the spam link is now gone (I have < 10k reputation on Stack Overflow).
@SmokeDetector the user name was previously associated with upmarketresearch.com. The answer is pretty useless; you would have to find the individual software publisher within HKLM_LOCAL_COMPUTER\SOFTWARE to actually do anything useful there. Is this ChatGPT-generated?
google.com/… brings me a large number of pages which seem to use hacked forums to discuss missing game DLLs ... not something I'm eager to click
@cocomac Thanks for the ping. Yeah, that's something that's started happening fairly recently and is, fortunately, rare. At the moment, I'm not yet sure why it's happening, as even scanning those an additional time should be prevented (it did used to be prevented and is still prevented under most conditions). There haven't been any changes in SD in that area since early December. So, it's possible that we're seeing issues with the SE API and/or fetching data from MS. The thing that's come to mind as a possible cause is that, maybe, the SE API is ping-ponging between slightly different versions of those posts, which would result in them being scanned multiple times, but if the versions were actually different, then I'd expect a new post record on MS when it's reported. So ... not sure yet.
Potentially more likely is an edge case or multi-threading issue which isn't being accounted for correctly in the recently scanned posts code. It is, however, unclear to me why we would only be seeing that now, given that it was last changed back in May.
My guess about it being related to the SE API is because we don't have proximate changes in SD's code that would easily account for the issue appearing now and the SE API is external to us and it's been demonstrated that even automatic changes/optimizations within SE's handling of SE API requests can make a significant impact (up to and including nearly crashing SD). It's also possible there were actual changes made in the SE API about which we are unaware.
It is, obviously, a problem and we need to look into what's actually causing it and resolve it. In the meanwhile, when it happens, it should be possible to get SD out of it by doing one of: giving FP or NAA feedback, telling SD to ignore one of the posts, or rebooting SD.
