Shut down website #743
Shut down website #743
Conversation
As-per [this asana ticket](https://app.asana.com/0/1207142155269470/1207142155269475/f) move the contents of `website/source/docs` to `/docs`
We may wish to keep this information (it is markdown after all!) and some of the docs reference it
As-per [this ticket](https://app.asana.com/0/1207142155269470/1207142155269473/f). The current content of these files are (were?) served up by the ruby application in `/website` and could be accessed using `/docs/filename.html`. Since this is being retired, this content now needs to be accessible (and navigable) from the GitHub repository alone. In order to accomplish this, links to `file.html` need to be rewritten to `file.html.markdown`. The suggestion in the ticket was to use a codemod to do this - however this could instead be accomplished using the following `sed` expression: ```bash find . -type f -name '*.*' -exec sed -i -E ' s|\/intro\/(.*)\.html\)|\/docs\/intro\/\1.html.markdown\)|g; s|\/docs\/(.*)\.html\)|\/docs\/\1.html.markdown\)|g' {} +; git status ```
As-per [this ticket](https://app.asana.com/0/1207142155269470/1207142155269482/f) the wbesite (and it's associated code) is being retired.
Co-authored-by: Byron Wolfman <[email protected]>
We should be linking to the relative URL, not the blob URL. Also, the file extension should be `.markdown`, not `.html` Co-authored-by: Byron Wolfman <[email protected]>
The correct link needs to have `/tree/master` in the URL, not just `github.com/serf/docs`. Co-authored-by: Byron Wolfman <[email protected]>
|
@skpratt - would you mind taking a 👀 . I think @byronwolfman's review was just to clean up the go linter errors for the most part. If you unassigned yourself because GH picked the wrong person, let me know and I can find someone else to take a look - cheers! 😄 |
rboyer
requested review from
zalimeni and
dhiaayachi
and removed request for
skpratt
|
Will the parked domain have content specific permanent redirects left so permalinks go somewhere equivalent? The consul and nomad websites deep link to the serf.io website: Specifically:
|
|
@rboyer I believe the parked domain will simply re-direct |
Confirmed that hc-centralized-dns parked domain capabilities are a simple "entire domain" HTTP 302 redirect; we don't have the capability to handle complex/per-page redirects with that mechanism. |
|
Are y'all going to take care of the followups in Consul and Nomad to avoid hitting the destination-removing 302 redirect for existing serf.io links? |
|
Here is a PR for the consul side: hashicorp/consul#21797 |
|
@jmurret Just FYI - I took a quick look at that PR....I believe direct file links need to contain |
|
thank you @rmainwork . i was planning on test links once your PR was merged. this helpful to know. |
|
Nomad site PR here: hashicorp/nomad#24114 |
The serf.io site is being taken down, so change all our links to point to the repo docs instead. Ref: hashicorp/serf#743
The serf.io site is being taken down, so change all our links to point to the repo docs instead. Ref: hashicorp/serf#743
This PR gets the repo ready to shut down the website (serf.io). However, I do have a few clarifying questions I'd like to call out.
docs/internals/simulator.html.erb..html.markdownas well? It does look like an empty template that something is supposed to fill in, presumably with graphics of sointrodirectory under/docs/intro(updating paths as needed) since some of the documentation under/docslinked to it, and it seemed like it might contain important context for people digging in to the project for the first time. Is this okay, or should I remove it?