Ours was almost completely unreachable for about 20 minutes, and is now mostly back online, except that ~10% of requests hang for ~5-15 seconds before being fulfilled. (This time is not accounted for in the 'connect' or 'service' times in the heroku/router log line.)

Yep. Started 21:56 UTC / 14:56 Pacific for us.

It's sad to see, because they used to be fully on top of stuff like this.

They finally posted to their status page about it: https://status.heroku.com/incidents/2451

> Beginning at approximately 21:54 UTC Aug 15th, routing availability for US Common Runtime-hosted apps dropped below acceptable levels. Engineering was notified and immediately began scaling up capacity to address demand needs.

I wasn't affected by it but I have been getting random R99 errors when launching cron jobs. About 0.1% of jobs are affected and fail to launch.

Support basically has no idea why. Flakiness without understanding why is somewhat worrying to me.

Same here.

https://twitter.com/yongfook/status/1559302434111234049

EDIT: Our app seems to be back up (2022-08-15 22:29:17 UTC).

Yup. Same here. totally down.

Yup, we're down too.

Yes, partial outage.

Same.