[16:13] <foo> We have a system that processes a lot of webhook events. If we have any server issues, at all, we lose those webhook events. We're processing financial transactions. This system needs to be bulletproof. I'm trying to think through the best way to do this. Any suggestions on how to commonly solve this problem? Hosting at digital ocean. The main issue we've seen, at least recently, is we had a memory 
[16:13] <foo> bottleneck which caused HTTP requests to hang. That hanging resulted in missed transactions. It's almost like we need another system that captures all webhook events, stores them in the database, and then sends them off for processing... and that system needs to be silo'd so it *only* captures endpoint traffic. Curious on thoughts here, thank you!
[16:17] <patdk-lap> I thought that was called haproxy
[16:30] <foo> patdk-lap: not familiar with that but can look into it :) 
=== elastic_dog is now known as Guest3315
[21:55] <cryptodan_mobile> What can cause this error 
[21:56] <cryptodan_mobile> Oct 24 21:14:10 capricorn agetty[1277]: tty1: invalid character 0xc in login name
[21:56] <cryptodan_mobile> I recently updated and went to use tty and all 6 of them were hosed and no keyboard input was detected. 
[21:58] <genii> Apparently it thinks you're sending a formfeed character
[21:59] <cryptodan_mobile> I ordered another keyboard so hopefully it's a faulty keyboard