Having issues with nodes in our scylla cluster randomly dropping out and then taking forever to rejoin, resulting in a bunch of degraded performance for the duration. RE: View quoted note →
If this whole non-archival relay thing moves forward as is, it means pretty much any dev can run a relay. Requirements move down to ~8 cores, 16gb ram, ~2Tb ssd (can reduce this too) and as much bandwidth as you want to provide to your downstream users Obv would scale with network, but a good start
firehose consumers may experience some issues for the next ~20 minutes, apologies, working on getting it back in a good spot.
To be clear, we do have plans for scaling, we just kinda expected more than a couple days notice before getting blasted with a million new users a day. The team is rapidly deploying fixes and new software to adapt. More servers in the mail.
Today has been a 'kick things until they start working again' type of day
Working on a feed of trending links, its not personalized yet (working on that!). But if you wanna see the most popular links people are sharing for now, you can check this out: RE:
honestly pretty frustrated that the outages today weren't even our fault. Makes for a lot of stress with no outlet except "buy more internet connections"
ISP problems this time, we even have two of them.
Features users want: - group DMs - custom notification filters - something about gifs Features we are prepared to deliver: - servers not going down under the load of most of twitter joining - some cute animations when you expand a picture in your feed
If you're running 20 firehose consumers on the same box in GCP please switch over to jetstream. Also just in general, if you're doing something with the firehose and don't *need* to be verifying every signature, please give jetstream a try.