HA for haskell.org

[ketzacoatl]

ATM haskell.org is down due to a hardware failure. It's been a while, it depending on the type of failure, could be a while longer before it is back up and available again.

Is there a way we can setup the haskell.org and its sub-sites (not the wiki, but cabal/GHC/ghcup docs, etc) across multiple servers so a hardware failure can be dealt with while haskell.org is still up and available?

Perhaps this is as simple as hosting the site on github's infrastructure, or configuring the CDN a certain way?

I'd be happy to help with the tasks/labor to do the setup for this.

[tomjaguarpaw]

In principle we could do that. There are a number of questions to resolve.

• Which hosting service would we use?

• How would we combine the subsites into one deployable artifact?

  Currently the subsites live as files in separate directories on the server. It's all a very much 20th century nuts-and-bolts setup. We could combine the subsites into this repo. That might have the beneficial effect of making them more easy to maintain.

• Would it actually make the website any more reliable?

  I don't know the uptime statistics for GitHub pages, CloudFlare, etc.. One's ability to convincingly demonstrate that sites served by them have less downtime than the current setup is directly proportional to our probability of switching.