* delete loads of really important stuff that we definitely need
* v basic MVP smoosh docusaurus PoC into repo
* TODO
* delete more really important stuff
* TODO
* tidyup: use run-s
* don't redirect images used in frontend to raster proxy
* fix routing
* preserve the /endpoint link
* delete the blog (for now)
I would quite like to re-add this at some point
but its not really the top priority thing right now
* content edits
* appease the lint gods
* update danger rules
* remove placeholder
* cypress tests
* dockerhub --> ghcr
* Revert "dockerhub --> ghcr"
This reverts commit ef74cbb26b.
* downgrade lockfile format
* implement defs/BASE_URL
* fix e2e build
* actually fix cypress tests
* always run cypress tests on build
* this never worked
* add command for docusaurus:clear
* delete more code we don't need any more
* update ESLint/prettier config
* delete unsused exports
* documentation updates
* delete a fairly large chunk of our dependency tree
* allow base_url as build arg to Dockerfile
* fixup dockerfile
* work out base url at runtime if not set
doing this at image build time is not the right approach
* remove gatsby monorepo from closebot
* rename HomepageFeatures to homepage-features
* Test server running with metrics enabled
* Verify custom metric was send
* Extra code removed
* Tests moved to core/server/server.spec.js
* tickAsync value based on push interval
* Trigger Build
Co-authored-by: repo-ranger[bot] <39074581+repo-ranger[bot]@users.noreply.github.com>
This is the code part of #3027, following [this article](https://www.viget.com/articles/heroku-cloudflare-the-right-way/) and using [this middleware](https://github.com/clive-io/cloudflare-middleware).
I pulled in the `addHandlerAtIndex()` function @chris48s wrote for #5574.
The middleware isn't perfect for scoutcamp, since it relies on `req.ip` which is something set by Express. However, the other solutions I found were either explicitly deprecated ([cloudflare-ip](https://www.npmjs.com/package/cloudflare-ip)) or relied on dynamically fetching the list of Cloudflare hosts ([cloudflare-ips](https://www.npmjs.com/package/cloudflare-ips)), which seems unnecessary as this list has not changed in several years.
I've left this off to start, so we can test it in production using an env var before we make it the production default.
* prom-client JSON to InfluxDB line protocol converter
* Converts a metric with separate names
* prom-client JSON to InfluxDB line protocol (version 2) converter
* Server has instance id
* Read the instance id from an environment variable
* More unit tests for instance-metadata
* Log instance id
* Push influx metrics
* INSTANCE_ID with dyno metadata
* Prepare influx metrics in one place
* Influx metrics endpoint should return metrics
* More readable tests
* Env added to instance metadata
* hostname as an instance label value
* HEROKU_DYNO_ID as an instance id for heroku
* Instance env can be set by env variable
* HEROKU_APP_NAME as an instance env
* Log instance metadata as a JSON
* Typo fix
* Code refactoring in tests
* wait-for-expect dev dependency added
* Test for pushing metrics
* Test for pushing metrics
* Use basic authentication for pushing metrics
* intervalSeconds=2 for development env
* Using existing methods
* TODOs removed
* Schema for influx credentials
* Influx config removed from config files
* Require username and password when influx metrics are enabled
* Unused args removed
* pushing component should log errors
* Speed up tests
* should log error responses
* InstanceMetadata class replaces by simple object
* Influx metrics can be configuredd by env variables
* Use application label name instead of service
* Unused code removed
* Integration test for prom-client and converter
* metrics.influx.enabled configuration option added
* Improved influx configuration schema
* instanceMetadata validation
* Typo fix
* Default value for env
* metrics.infux.hostnameAsAInstanceId added
* should add hostname as an instance label when hostnameAsAInstanceId is enabled
* Default values for influx configuration
* flatMap is not available in Node.js 9.4
* Env vars removed from Procfile
* Better instance metadata values in tests
* Typo fix
* lodash.groupby added to prod dependencies
* Allow other keys in private config
* Missing test - should allow other private keys when influx metrics are enabled
* Missing test - should require private metrics config when influx configuration is enabled
* log.error instead of console.log
* metrics.influx.uri -> metrics.influx.url
* Unused arguments removed
* async removed
* promisify sendMetrics
* Allow to disable prometheus metrics
* Create test server with custom config
* 'metrics-influx' resource removed
* 'metrics-influx' resource removed
* Private config schema flattened out
* Extra code removed in Prometheus tests
* promisify moved outside of the class
* Do not throw errors from got in a specific test
* hostnameAliases added
* instanceIdFrom added
* instanceIdEnvVarName added
* envLabel added to schema
* instanceMetadata is not used by InfluxMetrics
* Instance metadata removed
* hostnameAsAnInstanceId removed
* A comment added
* waitForExpect removed
* Unused code removed
* Changed configuration validation to allow using a named pipe for port.
* Update core/server/server.js
Co-Authored-By: Caleb Cartwright <calebcartwright@users.noreply.github.com>
* Applied prettier.
* Update core/server/server.spec.js
Co-Authored-By: Caleb Cartwright <calebcartwright@users.noreply.github.com>
* Updated spec for port setting to allow only numbers or formatted named pipes.
* Try enable universal cors
* Move handle from registerErrorHandlers because it isn't and error handler
* Add test for cors headers
Also add link to issue
This will definitely save time, and ensure more uniformity.
It moves the `createServiceTester()` calls to a different place from where I'd like them, though I'm happy to have them checked by the linter.
Closes#2701
While Next.js can handle static sites, we've had a few issues with it, notably a performance hit at runtime and some bugginess around routing and SSR. Gatsby being fully intended for high-performance static sites makes it a great technical fit for the Shields frontend. The `createPages()` API should be a really nice way to add a page for each service family, for example.
This migrates the frontend from Next.js to Gatsby. Gatsby is a powerful tool, which has a bit of downside as there's a lot to dig through. Overall I found configuration easier than Next.js. There are a lot of plugins and for the most part they worked out of the box. The documentation is good.
Links are cleaner now: there is no #. This will break old links though perhaps we could add some redirection to help with that. The only one I’m really concerned about `/#/endpoint`. I’m not sure if folks are deep-linking to the category pages.
There are a lot of enhancements we could add, in order to speed up the site even more. In particular we could think about inlining the SVGs rather than making separate requests for each one.
While Gatsby recommends GraphQL, it's not required. To keep things simple and reduce the learning curve, I did not use it here.
Close#1943Fix#2837Fix#2616
This test is being weirdly flaky in #2809. The problem seems to be in the test helper code, so I rewrote this using Joi.
I imagine the change has to do with a change to the test ordering. It's a bit puzzling.
However, the new test seems fine (and the endpoint is rarely used; not critical to begin with).
