It's not an IPv6 address, it's a block of IPs in CIDR notation: https://www.ultratools.com/tools/ipv6CIDRToRangeResult?ipAddress=2001%3Adb8%3A%3A%2F64 Blocking a single IPv6 address isn't very effective. https://github.com/NodeBB/NodeBB/issues/5879

@julian Nope. Found the post in my notifications. But the link also could have been in Google or my mailbox. @baris thank you for the fixing this extremely quickly!

@julian said in Drop Redis/Improve MongoDB/Support SQL?: Yes, let it be known that if we do decide to drop support, the Redis to Mongo migration script will be immediately published I figured as much, you guys are too cool to be jerks

I guess some backlash should be expected, new users won't suddenly come out of the woods, so you can only lose some of your active users in the first period after the conversion. But can we expect 5%? 20% 50% decrease in traffic? And what's to be expected of the situation a few weeks or even months after the conversion? We would move because we think users will stay engaged more with notifications, mentions, modern looks, better usability. We also expect new users to sign up in bigger numbers because this will be easier (SSO) and also because of the improved usability and better looks. But that's all theory so far. Some real life stories about conversions of big boards would be most welcome

I get your drift But that's not quite what I meant. My proposal would be to save approximately what a user enters in a textbox on the site. I.e. the same that happens now. So you'll save <b>blabla</b> instead of **blabla**, but you'll just keep NodeBB/NodeBB# 1234 as it is. When you edit the post <em> is converted back to * if the composer you use accepts Markdown. If the site uses a a WYSIWGY editor you can just keep the HTML and work with that. Plugins can still hook into all that, just make sure that they don't put presentation HTML in the DB. I can imagine that they do add something like <widget=github>NodeBB/NodeBB#1234</widget> to the text that goes in the DB (and deal with this appropriately when editing). This would make it easier to build more complex widgets I guess. But that's all optional. My main point is that saving (really basic) HTML in the DB is much more universal and more extendable than saving Markdown. Well, actually my main point is that I found a different (better?) way to deal with the realtime preview of the MarkDown