Skip to content

RackNerd DNS down, support silent for 72 hours

Domain Names by annexbi 15 replies 2.4K views
#11
alphadog said:
Which region did they mean?

They didn't. "Region" is placeholder for "we have no idea yet." I have seen this script at OVH, at Online.net, at every budget host since 2005.

The real question is: do they own their AS or rent it? Because if they are reselling someone else's DNS infrastructure, the actual fix depends on a third party they may not control.

#12

I moved my two domains off RackNerd DNS yesterday to deSEC. Free, API works, anycast from their nodes in Frankfurt and Hamburg. Propagation was under 10 minutes.

My RackNerd VPS in Chicago is still up but I only use it for non-critical stuff. €12/year, expectations calibrated accordingly.

#13
moebig said:
Transferring to HostHatch tonight

HostHatch is good VPS but they are not registrar, no? You cannot transfer domain registration there. You mean change nameservers to HostHatch, or you move registration to other company?

I use Gandi for registration and Vultr DNS for resolution — https://www.vultr.com. Separate since 2019, never problem like this.

Also I test again from Madrid, still timeout from RackNerd NS. 96 hours now for annexbi, this is very bad.

siesta first, deploy later
5 #14
annexbi said:
Tested from: my home connection (Contabo ISP)

Contabo in Bangkok routes through Singapore, at least on my line. Not relevant here—RackNerd's NS are dead from everywhere.

I have 4 domains with them, all on their free DNS. Lesson learned. Moving to ClouDNS free tier tonight, they have POP in Bangkok so latency is nice.

RackNerd support still silent on my ticket #48344 too. 54 hours.

#15
olespete said:
Already seeing reports on Twitter that RackNerd's whois server is also timing out

Confirmed from Toronto. Whois.racknerd.com:43 connection refused. Their RDAP endpoint at rdap.racknerd.com is returning 503 with a Cloudflare ray ID, which means origin is down and CF is serving stale.

This is not "degraded DNS performance." This is control plane failure. The DNS, whois, and RDAP stacks share nothing technically except the people who run them and possibly the network they sit on. If all three are down, you are looking at either cascading failure (bad deploy?) or network partition at their edge.

I would be checking my registry locks if I had domains there. Not saying hijack, but saying the conditions where hijack becomes easier are present when your ability to monitor is impaired.

It's always DNS. Always.
#16
FlowSana said:
I would be checking my registry locks

How? RackNerd portal is up but auth codes page gives 500 error. I cannot even start transfer to another registrar.

My domain is .com so I wait 60 days anyway from registration, but I want to prepare. This is hostage situation.

Post a reply

You need an account to reply. Log in or register to join the conversation.

Post reply Preview Save draft