And it is thought, iis to leave as and a reverse-proxy with SSL.
You could not impart experience more in detail?
Interests as you implement balancing and how you implement life cycle of servers?
For example it is necessary to update the version, you stop service, copy new bins, then launch service?
Whether there is at you a similarity of web gardens for fail safety?
At us nginx as a Web server.
Web applications on asp.net core are published in a type of services and dockers-containers.
Completely 100 % all on Octopus, server groups for seamless release are divided on tenants, packets with releases are allocated local nuget repositories, in the same place units for the assembly.
When it is necessary to update the version, we come in Octopus, we push, Create release, the channel of delivery and we press button Deploy. When half of machines , turnes on the mechanism of transfer of users on new machines when they are thrown, all connections are closed, the trigger and either in manual, or in an automatic mode is launched, happens on a remaining half.
There are mechanisms as all at once, the full automatic machine. And it is partially possible subsystems, services, services separately. All seamlessly.