Ask HN: Where do you deploy to in 2018 and how? • Hacker News Board

Re: Ask HN: Where do you deploy to in 2018 and how?

#61

Post by rsto » Tue, Feb 27, 2018, 3:42 PM UTC

PerfectElement wrote: post #57

I do things in an old-school way. 2 Bare metal servers at different data centers with totaluptime.com acting as a load balancer with auto-failover. Deployment is done via SFTP by pressing the publish button in Visual Studio. This will deploy to the inactive server. I then manually trigger tests on GhostInspector (this could be automated via API) to make sure I didn't break anything. Then I run a custom script to make…

Thanks for sharing, I'm considering bare metal as well for a project. Is latency between the two data centers an issue for you, e.g. is one of your two servers running a SQL database as master?

Re: Ask HN: Where do you deploy to in 2018 and how?

#62

Post by jesses » Tue, Feb 27, 2018, 3:45 PM UTC

If you're using Elixir, take a look at gigalixir.com I'm the founder.

Re: Ask HN: Where do you deploy to in 2018 and how?

#63

Post by superkitty » Tue, Feb 27, 2018, 3:52 PM UTC

Mesosphere DC/OS, AWS, ECS, EKS, EC2, cloudformation, Jenkins, deployment API developed at home grown...

Re: Ask HN: Where do you deploy to in 2018 and how?

#64

Post by superkitty » Tue, Feb 27, 2018, 3:53 PM UTC

What folks use for deployment strategy for machine learning based models?

Re: Ask HN: Where do you deploy to in 2018 and how?

#65

Post by yamalight » Tue, Feb 27, 2018, 4:04 PM UTC

Have been rolling my own server on Scaleway / Hetzner and deploying with Exoframe [1] for past ~year. Works pretty well :)

[1] https://github.com/exoframejs/exoframe

Re: Ask HN: Where do you deploy to in 2018 and how?

#66

Post by rickyspanish » Tue, Feb 27, 2018, 4:30 PM UTC

PerfectElement wrote: post #57

I do things in an old-school way. 2 Bare metal servers at different data centers with totaluptime.com acting as a load balancer with auto-failover. Deployment is done via SFTP by pressing the publish button in Visual Studio. This will deploy to the inactive server. I then manually trigger tests on GhostInspector (this could be automated via API) to make sure I didn't break anything. Then I run a custom script to make…

This is probably a very dumb question, but what exactly do you mean when you say you run bare metal servers at a data center?

I run a kind of similar setup sans Visual Studio, so I'm very interested in understanding your setup a little better.

Re: Ask HN: Where do you deploy to in 2018 and how?

#67

Post by neurostimulant » Tue, Feb 27, 2018, 4:39 PM UTC

Rancher is actually pretty solid. The API is actually pretty good, allowing me to write some automation scripts for everyday tasks.

Re: Ask HN: Where do you deploy to in 2018 and how?

#68

Post by Jean-Philipe » Tue, Feb 27, 2018, 4:41 PM UTC

Try `now` from https://zeit.co, it's super pleasant. Hacker friendly. Though, mostly nodejs centered (AFAIK).

Re: Ask HN: Where do you deploy to in 2018 and how?

#69

Post by sergiodxa » Tue, Feb 27, 2018, 4:54 PM UTC

Jean-Philipe wrote: post #68

Try `now` from https://zeit.co , it's super pleasant. Hacker friendly. Though, mostly nodejs centered (AFAIK).

It supports Dockerfile so you can deploy any backend technology using Docker containers, not only Node.js

Re: Ask HN: Where do you deploy to in 2018 and how?

#70

Post by potta_coffee » Tue, Feb 27, 2018, 5:43 PM UTC

krambs wrote: post #48

surge.sh is stupid-simple for static sites.

I love surge, it's been awesome.