It’s all domain name: irc.twitch.tv or irc.chat.twitch.tv
(whole thing with Amazon AWS is that they can add/remove chat servers at will without telling us)
but why now is possible to connect and get users lists only?
The old clusters are still up and have userlist information even though messages aren’t being transmitted anymore. They won’t be up for much longer though.
close it faster 
but you should transmitt messaged until end. This IP was from the beggining, worked even when irc.twitch.tv had problems. So naturally for me was given higher priority over irc.twitch.tv. How could i know that in the future it will be still possible to connect to this server and not get messages
Seeing an unusual amount of connection failures and/or connection drops on AWS. Not sure if just me?
I have been connected to AWS for 30+ hours without issues on any of my connections. Might be something on your end?
Are you logging failures/timeouts when opening a connection? That’s what I’m seeing more of.
Ah, yes. It failed a couple times when it tried to connect. I’m not having that issue now though.
Been green on my side (Note: I don’t connect to whispers, only aws main chat) - Only one connection has dropped today, and no servers have been added to my block list today. (Failed connections block for 5 minutes to wait for the error to clear).
I am pretty sure I have a big enough sample size to be relevant to moobot type connections.
Looks like I had a mass exodus of connections yesterday though - but no alerts about ips being blocked, just connections being closed. those were the main / event server connections shutting down due to no recent ping/message.
Theres only one server though right?
So if you get a connection drop, you’d lock your self out?!
There’s one DNS, which I resolve to an IP address. Each IP address is treated for whether or not it needs to be blocked.
Ah so you’ve managed to get a list of IP’s that exist.
I wonder if the connection drops are related to Twitch adding/removing chat servers to the cluster shrug
The connection shouldn’t drop once you’re connected, but if you don’t have the most recent ip list and they’ve turned off listening on that server you may be getting timeouts by the dns resolving to an old IP. I made sure to make all of that white noise on this iteration though, with the consolidation of the aws endpoints I was able to clean up my connection code a lot so there wasn’t a bunch of branching on ‘if x, y, z scenarios are true, do a, b, c’ - and instead it can focus solely on connect to a cluster server, and transparently migrate between available back ends based on the returned aws servers resolved.
As a FYI for anyone else here having issues today, or Twitch staff. About 30 minutes ago Revlobot started disconnecting from all channels and was unable to connect anymore when we restarted servers.
Checking: http://tmi.twitch.tv/servers?channel=revlo_adam we were getting servers with darklaunch in it and couldn’t connect.
irc.darklaunch.us-west2.twitch.tv:80
After hardcoding irc.chat.twitch.tv into our app we were able to connect fine again instead of using the darklaunch from tmi. We’re still investigating more on our side to see if we’re doing anything weird connecting, but it looks like that may have been an incorrect DNS to connect to.
I’m seeing the same on all the channels I’ve tried. Something is definitely broken.
Should be fixed now.
still online -.-
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.