Amazon Cognito now supports multi-Region replication | Hacker News Reader

Amazon Cognito now supports multi-Region replication

aws.amazon.com

30 pointsmooreds17 comments

i try to keep my comments on here positive, but man, my experience using this product has been awful.

You should really stay away from any "high level" AWS product, as there is almost always something much better and more flexible.

Use EC2, EBS, S3, Route53 plus EKS, RDS, ElastiCache.

But anything else that isn't low-to-mid-level (looking at you Beanstalk), use something better.

Clerk, WorkOS, etc.

Cognito isn’t a high level product not anymore than any of those you’ve mentioned it’s a CIAM solution.

It's a Clerk/Auth0/WorkOS alternative.

If you implemented it yourself, you would do it on top of those.

This has been an obvious feature request for many years, but glad to see AWS investing in what started to feel like a service that was mostly abandoned for investment.

I recall complaining about this with one of my architects who was looking to implement Cognito round about 2019.

Cognito just supported multi-region? For identity this seems like a very high priority issue. I was at a company 10 years ago that we didn't use Cognito to build, and build our own AWS based identity because Cognito didn't have this (and just seemed pretty half-baked).

They wanted to rebase onto a different database first to make multi-region easier, but that work took many years.

This should have been available from the beginning. I don't understand why it took so long.

I think cognito was internally low-staff/KTLO for a while and that changed recently.

What does KLTO mean?

Probably meant KTLO: Keep The Lights On

"Keep Lights To On." It's the post-it on the light switch wired to the Cognito server.

To add to the other posters, keep-the-lights-on usually means a product has no active feature development. It’s just supported with on-call and maybe some bug fixes depending on capacity.

No clue if Cognito actually was KTLO though.

This prevented us from failing over during last October's outage (unless we wanted to reset everyone's password). Glad to see AWS focusing in on resiliency.

I work for a Cognito competitor, but I am glad to see them investing in improving the lives of folks using this native AWS service.

It felt like Cognito was abandoned for a while.

Yay, it's only taken them years to do this.

Since the pool identifiers are static, how do you actually fail over?

Oh, you need a custom domain that presumably routes if the primary dies.

Reply on news.ycombinator.com