Recovered from ssl.SSLError: [SSL: BAD_LENGTH] bad length (_ssl.c:2130)
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, bad keyword in title, blacklisted website in body, body starts with title and ends in URL, +2 more: Turmeric Forskolin by user88549 on workplace.SE
@bertieb I'm not sure what you're asking. Are you asking if the reasons "Bad NS for domain in X" have normally determined weight or that the weight is forced to be 1 like the "Potentially bad keyword in X" reasons?
You probably chose them because they are cheap, easy, and/or have minimal requirements. That's attractive to spammers too. It's not really a comment on the quality of the NS. It's just a correlation that a significant number of the spammers that put their domain in spam on SE tend to use them.
I kinda-sorta thought that was what you were talking about, but wasn't sure. There's no way for us to know the actual problem, because the user didn't say what it was they wanted to be in there.
@Makyen That pattern looks like it's already caught by Pattern-matching website in answer and Pattern-matching website in body; append -force if you really want to do that.
Hmmmm... I'm not going to watch that. While I think that it should be something we keep an eye on, the pattern recognition detections will be sufficient to give us notification of it's use. Watching it in addition to those would tend to indicate we've seen it in obvious spam, which the above post was not.
Why exactly was 6 flags chosen as the threshold to delete spam posts and penalize the author with -100 reputation? Was 6 calculated to be the optimal number?
File "/usr/local/lib/python3.6/dist-packages/websocket/_app.py", line 118, in send if not self.sock or self.sock.send(data, opcode) == 0: File "/usr/local/lib/python3.6/dist-packages/websocket/_core.py", line 234, in send return self.send_frame(frame) File "/usr/local/lib/python3.6/dist-packages/websocket/_core.py", line 259, in send_frame l = self._send(data) File "/usr/local/lib/python3.6/dist-packages/websocket/_core.py", line 423, in _send return send(self.sock, data) File "/usr/local/lib/python3.6/dist-packages/websocket/_socket.py", line 116, in send
> Bad keyword in body, bad keyword in title, bad keyword in username, pattern-matching product name in body, pattern-matching product name in title ---------- Title - Position 1-17: male enhancement, Position 18-34: male enhancement Title - Pattern-matching product name *male enhancement, male enhancement* Body - Position 1-17: male enhancement, Position 18-34: male enhancement Post - Pattern-matching product name *male enhancement, male enhancement* Username - Position 1-17: male enhancement, Position 18-34: male enhancement
[ SmokeDetector | MS ] Blacklisted website in body, body starts with title and ends in URL, link at end of body, potentially bad keyword in body, potentially bad keyword in title: Keto Complete you may be capable by user45369 on askubuntu.com
Options are parsed from left to right, and stops at the first invalid option, so asdf site=foo will be tested against blacklists instead of parsed as options.
@SmokeDetector @undo Did you get the logs working? If so, here's another drupal.se autoflagged post with only 3 autoflags, but high weight. That's at least 4 posts in the last 24/36 hours where this has happened.
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, blacklisted website in body, body starts with title and ends in URL, link at end of body, +3 more: joined in this novel recipe! by Wornat57 on askubuntu.com
[ SmokeDetector | MS ] Bad NS for domain in body, blacklisted website in body, body starts with title and ends in URL, link at end of body, pattern-matching website in body: Keto Fit Calorie as components by yrvbna12 on askubuntu.com
Recovered from ssl.SSLError: [SSL: BAD_LENGTH] bad length (_ssl.c:2130)
[ SmokeDetector | MS ] Bad NS for domain in body, bad keyword in body, blacklisted website in body, body starts with title and ends in URL, link at end of body, +3 more: We try on obliged this snappish by Ablemody on askubuntu.com
The toxic post and offensive post reasons usually end up together, but there are cases where there's toxic without offensive, and offensive without toxic
> Bad keyword with email in body, bad keyword with email in title ---------- Title - Obfuscated email dr_mack@ yahoo. com Post - Obfuscated email dr_mack@ yahoo. com
@TheMaskedRebel You are not a privileged user. Please see the privileges wiki page for information on what privileges are and what is expected of privileged users.
@Undo I can also try to do helios stuff, but I'm not totally sure what that entails. I think it's bascially building a front-end for helios (to view stuff) and building a system where SD can send pending BL stuff to MS, then MS can get it approved and send it to helios.
Ok. I could make an issue on MS with my thoughts? Then I can build out the MS stuff, helios is built, and then someone else would have to make the SD changes.
@iBug It depends on what you're trying to get. If the message has been edited, then it's easier to just click the link for the message's history. If it hasn't been edited, then you need to construct the URL, one way or another. Obviously, if you're attempting to parse it programmatically, it's easier to get just the plain=true, if available.
@iBug I'd, of course, appreciate that, but given how rarely it happens, and that there's an easy work-around (just don't edit SD commands you really don't want to repeat (i.e. a !!/watch that's been applied)).
I've seen a lot of SD command messages edited. That one was the only one I've seen where SD picked up the unedited version, rather than the new version.
I, and others, routinely edit !!/watch (or blacklist) into !!/watch-force when SD responds that the string is already caught for some reasons which don't make it inappropriate to watch/blacklist. That specific time was just very inconvenient because I was editing a !!/watch which had actually been added into something slightly different. Because it took the original version, it added two identical watches, which breaks CI and requires someone with GitHub write access to fix.
So, the workaround is to just not edit when repeating the same command again would cause problems (which is easy, now that we know it might be an issue). Yeah, it's rare. I've only seen it once out of at least dozens of times I've edited, or seen people edit.
I also have a 6 month out of date version of smokey that talks to Helios. No idea what it's going to take to get it up to date with the current version though
@Andy Ok. I'm about to write up an issue in Helios, and after that you can figure out if it'll be more work to merge current SD into your 6 month old fork or just start over.
Why exactly was 6 flags chosen as the threshold to delete spam posts and penalize the author with -100 reputation? Was 6 calculated to be the optimal number?
