[02:23:37] more of these damn things https://meta.miraheze.org/wiki/Special:AbuseLog/1134602 https://meta.miraheze.org/wiki/Special:AbuseLog/1134587 [02:23:57] @kiju1108 these spam things even worth a CU? [02:24:03] all the previous ones came up blank [02:26:20] I think we should consider setting filter 18 to disallow again [02:26:29] i dont see it firing any false positives [02:27:12] actually the external link one covers most of it [02:27:27] seems they get around the first oen actually [02:27:44] https://cdn.discordapp.com/attachments/443926951292567562/1281803058803638392/TdneByg.png?ex=66dd0b9f&is=66dbba1f&hm=2a452907e77bac8ca616aaa389d7429ec1c7f07a3b858d8760f74f2525f7c72b& [03:03:54] Okay. will CU [03:11:53] If an account hasn't actually made any edits and is just being blocked by AbuseFilter, it's harmless, so there’s no point in running CU. It would be endless [03:14:06] There are cases when it's warranted if there's multiple attempts and some are getting through [03:14:29] But generally not worth our time to chase, yeah [03:14:59] The rule I used to apply was to CU random spam accounts and check their /24 or entire ISP for signs of malice [03:15:29] Sometimes I would find entire ASNs which were bot farms and so I would promptly block it [03:15:34] but this only works if you have time [03:26:10] Yep, this bit being the most crucial - time is our most scarce resource, so not an everyday maintenance activity. [03:27:00] [1/2] if we have time, definitely worth doing [03:27:00] [2/2] could be some room for partial automation [03:27:16] certainly not [03:27:26] you should only do this at 3am [03:27:37] when you should be sleeping \:O