I have reservations about outright blocking by IP. But …
Even though the number of IPs that can be abused is effectively infinite. I have found it extremely helpful using Registration IP to identify possible and probable problem accounts. - I currently have over 180K associated with Banned accounts.
Most often the IP used to register is not the same IP used to post.
Unless I’m wrong, it seems that Discourse uses the Last IP. i.e.
if no posts == registration IP
else == posting IP
I strongly believe that having both would be of great benefit for moderating.
Any chance of getting this worked into the Core?
My thinking is we’ll be writing custom code to enhance moderation anyway, but we can only work with what information is available.
[quote=“Mittineague, post:1, topic:14884”]
Unless I’m wrong, it seems that Discourse uses the Last IP.
i.e.if no posts == registration IP
else == posting IP
[/quote]Not necessarily post IP; it seems to record last log-in IP, whether a post was made or not.
I added this today. It shows in admin below the last ip address:
Also, when an admin deletes a user and chooses to block their email and ip address from signing up again, it will also block the registration ip address if it’s different than the last ip address.
I’m certain that some of the other Mods eg. @TechnoBear and @cpradio will be as pleased as I
I’m not a SPAMmer so I don’t know for sure, but it seems like often someone or something Registers a mess of accounts using the same IP, then sells the accounts to others that use a lot of different proxy IPs to make posts.
There are an awful lot of common IPs abused for making posts, true, but I have found that Reg IPs are more accurate for spotting and identifying problem accounts so this is indeed great news.
It also usually helps identify someone using multiple accounts, or setup posts. It just gives us another tool to take into account when investigating suspicious topics.