@jflasher Where do you propose the status page to be hosted? gh-pages or S3? If the latter, could you create a bucket?

Not time-sensitive

@ascalamogna The status page is ready for your review.

I tried to keep as simple as possible and used the openaq-design-system to apply some styling.

This is how warnings on the Fetch section are shown and stacked:

Let me know if you need anything. To run status locally, you'll also need to run the API locally.

cc @danielfdsilva

Maybe add the link to the status pages (human readable one and API one) in the README/description of this repo?

For consistency let's migrate this repo to Github Actions

Hi there,

Is there currently a system outage? I haven't been able to get any data back from the v1 or v2 API for days (500 error response).

The functionality of the status page is implemented. We're still revising the style and layout, but it's ready for your review.
Most importantly, please review the statuses and feedback messages.

API health

Here we are fetching the status from New Relic. Their green, orange, red system is a bit ambiguous, so it's hard to come up with a very descriptive error message. For now we have:

API Fetch

On the API Fetch side, we are doing a couple of checks.

At the moment, there is nothing that really causes an error message on fetch.

@jflasher @RocketD0g Some feedback on this would be great. Anything else that we can determine from either the New Relic API or the fetches endpoint?

the calendar does not properly refresh when left open over a month turnover i.e March 31 to April 1. Likely need to redraw calendar on updates or check for month turnover

Getting 403's on the images.

For instances where there is maintenance that might not be accurately reflected in other sections, for example knowing the database will go up and down. This could be as simple as a section in the index.html that's easy to comment out when not needed.

API is providing proper response.

After fixing basic functionality it would be good to get Node to a modern version for ease of development

A ticket to discuss the overall scope of the status page. The first version will reflect:

1. API health
   New Relic: green, yellow, red
2. Fetch health
   Hit the fetches endpoint and check the last time(s) it ran. We can include some summary stats like: total measurements added from x different sources. Also total failures and their reasons.

In the future, if we want to do more complex analytics, for example on Source level, this could be added to Fluffy. This could include source health, by checking the amount of measurements reported by station.

The first iteration will be similar to the OAM status page.

cc @danielfdsilva @nbumbarger @ascalamogna @jflasher @RocketD0g

I'm seeing 500 and 503s on most endpoints (all except countries), but status.openaq.org reports green.

Do you want to track Analytics for the Status Page? If so, what is the tracking ID?

cc @jflasher