Tell HN: AWS Is Down | Hacker News Reader

Tell HN: AWS Is Down

10 points65012 comments

Was looking for this post. We've been having very strange issues since ~11:30 AM ET in us-east-2. We have a DB instance that became suddenly unresponsive to queries even though we could still shell in, almost like a disk died or something. Rebooting the instance has been hanging for >30 minutes. Seeing a smattering of other things not working as intended as well.

> [9:32 AM PDT] We are experiencing performance degradation for a small number of EBS volumes in a single availability zone use2-az1 in the US-EAST-2 Region. We are actively working on resolving the issue but don’t have an estimated time of recovery at this moment. Customers should restart from an EBS snapshot or failover to alternate availability zone in US-EAST-2 if the application supports it to maintain application availability as we continue to work recovery.

from https://old.reddit.com/r/sysadmin/comments/14n4mym/useast2_l...

Our DB just finally finished rebooting and is now responsive to queries again, so they're definitely making progress

https://health.aws.amazon.com/health/status says everything is fine though /s

downdetector says something different: https://downdetector.com/status/aws-amazon-web-services/

sad that you can rely on their health dashboard. doesn't add any value unfortunately.

See also: https://www.reddit.com/r/aws/comments/14n4ig1/useast2_networ.... AWS Support is out of sync with ground-truth, which appears to be a reliability issue.

Also, were are the post-event summaries from AWS[1]? Did AWS decidedly opt-out of post-event summaries and CVEs? Is that even aloud? This is worrisome.

[1]https://aws.amazon.com/premiumsupport/technology/pes/

Yep, it appears that I'm getting bitten by this one even though I don't use AWS, but a supplier does. I see posts about it, but I can't seem to find any public info on their dashboard, which would be nice to show customers.

Both Google and Microsoft report issues with Cloud services. It's astonishing that AWS does not. What is being covered up exactly, and for how long?

Widespread issues in us-east-2

Rumors say construction cut a cable.

Ah, the fiber-seeking backhoe strikes again https://imgur.com/a/Yqva7

Our vendor tells us their product hosted in AWS Central is down because of an outage there.

On a Friday! Feels for Engineers handling this across companies. :-|

Reply on news.ycombinator.com