Hacker News Reader

Comment by overshard | Hacker News Reader

As someone who hosts hundreds of PAID sites with CloudFlare this is pretty unacceptable. I'm giving them thousands of dollars so that this doesn't happen. Will probably be moving off unless they have some very good reasoning behind a world-wide shutdown of a geo-redundant service...

View on HN

Have you called the enterprise support number? What was their response?

We are in the same boat, but guess where I have the phone number for enterprise support? Webmail on our domain. Lesson learned.

The response is "sorry we are looking into it, we'll call you back when we know what's wrong"... pretty generic.

Well to be fair, what do you expect them to say before they have diagnosed the root issue?

Fair enough. But then what is the point of paying for enterprise support if all you get is the same generic response a normal punter gets? If they cant offer better, then enterprise support seems poor value at best.

Are we sure that if CloudFlare wasn't so popular here, the word scam wouldn't be used?

So you're saying you want them to say the same thing but reword it since you're enterprise? If they don't know the root cause, they don't know the root cause.

CloudFlare offers CNAME option for paid customers. So you can use an enterprise DNS service and only point to CloudFlare via a CNAME record.

When disasters like this happen, a quick DNS change can be a life-saver.

There's no such thing as a quick DNS change.

My startup NameTerrific can support instantaneous DNS updates in a geo-redundant Anycast infrastructure. As long as your TTL is sufficiently low (<300), the impact is quite limited as propagation time is negligible at NameTerrific.

EDIT: Sorry guys. We got some issues with a gem after installing the recently updated ruby2.0.0p0. The unicorn workers were timing out. TerrificDNS is completely unaffected and the site is already running again.

Not the best advertising

http://www.nameterrific.com

"502 Bad Gateway

nginx/1.2.7"

Hopefully the back-end is not running Ruby. ;-)

The back-end is AWS Route53, it seems.

Well, we have already soft launched our own TerrificDNS Anycast and it has replaced the Route 53 solution. TerrificDNS platform is running on Redis + PowerDNS.

Terrific.

It's up now. Please see parent for explanation. Sorry!

https://www.nameterrific.com/pricing

"We're sorry, but something went wrong."

From my experience with DNS for a large site over the last years, a lot of ISPs do not respect your TTL and have their own TTL (often 24h).

If and only if the site's end users and their software fully respects TTL. A lot don't, especially shitty mobile networks and web browsers.

You should probably put the same amount of effort into your web hosting infrastructure http://imgur.com/ok5lfml

Nameterrific is down. https://www.nameterrific.com/

Would love some details on how to do this, I've been a (business) CF customer for a while now and have never seen this service and I haven't turned up anything googling/scouring their website for it just now.

Details much appreciated!

EDIT: Spoke too soon: https://support.cloudflare.com/entries/22054357-how-do-i-do-... shame it's so convoluted

That is the first time i have heard this as well. Thanks for the Link.

Yup, this CNAME service just allowed me to make a quick update to DNS and get us back up. We will probably need to automate this in the future.

Yep, this is definitely something I'll do in the future and I guess it was my fault for trusting a single service... even one that is designed around keep your site up/redundant.

Reply on news.ycombinator.com