Respond quickly to possible service interruptions by knowing about them first.
If you depend on upstream systems, you want to know when they have problems, just like you know when yours does.
What does this have to do with anything ?
Most upstream systems have status pages …. and RSS feeds too !
On a side project I am currently working on, it is VITAL to know when upstreams start having hiccups, as we depend on their API’s as much as we do ours.
Unsplash API — If it’s down, we cannot serve images
Digital Ocean — If it’s down, we cannot update DNS for multi tenancy
Gitlab — If it’s down, we can’t build anything, nor deploy (If the pipeline doesn’t deploy it, nothing deploys it)
So we created a channel in slack called #status_feeds. If anything goes wrong, we know immediately.
On holiday ? You’re entire slack should be muted anyway so no worries !
Here’s how to add your own.
1. Pick an upstream you want to watch … constantly.
We’ll take Unsplash for this example.
2. Google “Unsplash status”
For the most part the first result should be the one.
3. Check on that page for a ‘subscribe’ button. Unsplash’s is fairly easy:
4. Click on the subscribe button and look for the RSS feed option
5. Click on the RSS Feed option. This will usually redirect your browser to a page with a bunch of XML on it.
6. Copy the URL in the browser
7. Go to your new slack channel for status feeds, and send the following message:
/feed subscribe [url] # In our case /feed subscribe https://status.unsplash.com/history.rss
8. Relax. Slackbot will let you know when your friends are having hiccups !
9. Now go add some RSS feeds to your APIs so other people won’t email you about an error. They will get a notification that you already know !
We also have our Sentry errors and CI / CD pumped straight into our slack.
If anything goes wrong, we want to know about it.