26 comments

[ 3.0 ms ] story [ 72.0 ms ] thread
Was down; back up (for me)
I am seeing issues with several services.

AWS might be a common link. I saw some CloudFront errors, but their status page is green top to bottom at the moment.

Heroku and Github have incidents on their status pages now, BTW.

Seems to be all good now. Any idea how long the outage was?
My application uptime monitor clocked it at 12 minutes.
Almost exactly 12 minutes for me as well.
(comment deleted)
WTF was that?!? Smells like some sort of reroute attack to me.
Can we please fix the title of this post and/or its link?

It currently takes you to Heroku's homepage, which contains no information relevant to the post's title.

I guess that answers the question, at least…
Just resolved it seems. Hit us in the middle of a huge enterprise demo. We’re moving off Heroku as fast as we can, but we need to make the jump to containerization first and it’s hard to prioritize infra projects these crazy days.
Seems like an overreaction. Moving to where?
We had approximately 13 minutes of downtime across all apps. Even Heroku's own docs were down
Seems up at the moment. I'm sure there will be some outages as data centers are running with reduced staff and some companies don't know how to do teleworking well.
Would be interested in hearing what heroku users are using for automated failover when heroku goes down.

Obviously can't live in the application code (b/c that's unreachable at that point), but is there an easy way to configure Cloudflare or something to switch to a static site hosted somewhere else?

Cloudflare itself will host a cached version (plus a little Cloudflare-injected banner at the top) in this situation, but not apparently not all of the time reliably. It has worked for us in other "520" error situations.
both my sites are working as of right now. Looks to be resolved
The outage seemed to be limited only to SSL requests. Since all non-ssl requests to .heroku.com are forcefully redirected to ssl, all .heroku.com requests failed during the outage.

We were able to handle requests at our http://*.herokuapp.com endpoint, but all https endpoints were down during the outage.

I had roughly ten minutes of downtime to due a certificate error between Heroku and Cloudflare. The issue resolved itself.
(comment deleted)