Skip to content
Register

5 years with RackNerd's $3 plan, somehow

Deals & Offers by jane_ffm 25 replies 2.7K views
jane_ffm

Member OP

#1

I know, I know — we're not supposed to admit we stay with budget hosts this long. But I need to share this somewhere.

My $3/mo "Bronze" plan with RackNerd has now run 1,827 consecutive days without a single unplanned outage. Not one. I've migrated clients off premium providers with worse track records.

  • 2 GB RAM / 1 vCPU (shared)
  • 40 GB SSD
  • "Unlimited" bandwidth (fair use, they say)
  • cPanel included
  • Ticket response: usually 4-8 hours

The funny part? Their status page hasn't updated since 2022. Their blog is a 404. Their WHOIS privacy expired. Yet my uptime monitor pings green every 60 seconds.

Anyone else riding a zombie host into the sunset? I'm half-convinced I'm their last paying customer and nobody's told me to leave.

— Jane @ Contabo

Single mode till I die 💀
brusselsdzire1

Member

#2

This raises fascinating questions under the Digital Services Act and GDPR Article 32. If RackNerd has effectively abandoned operational oversight, who is your data processor of record? Have they maintained a valid DPO registration in the EU?

I had a similar situation with a Lithuanian registrar in 2019. Article 13 would have been relevant had it passed in its original form. As it stands, I would request their SCCs immediately — if anyone answers the ticket.

mediaaustin

Member

Deliverability Nerd

#3

Your deliverability is probably suffering and you don't know it.

For what it's worth, a host with no operational staff means no abuse desk. No abuse desk means compromised neighbors sending pharma spam. That IP reputation degrades slowly — you'll only notice when Gmail starts soft-bouncing your newsletters.

Check your sending IP against Barracuda and Spamhaus. I'd bet money it's listed or teetering. SPF and DKIM won't save you if the /24 is poisoned.

SPF, DKIM, DMARC — holy trinity ✉️
web10

Member

off-grid admin

#4

23/04/2025

Bloody hell, 1,827 days is 5.003 years by proper calculation. I'm more impressed you've tracked it to the day.

The metric that matters: what's your actual load? My guess is you're running static HTML and a contact form. A single vCPU at 100% for thirty seconds would probably trip whatever ancient OpenVZ container you're stuffed into.

I've seen this before. Host oversells into oblivion, most customers churn, remaining accounts drift on autopilot. Not a business model. A statistical accident.

lee_mcr

Member

#5

Right then, I'll be honest — we've got accounts from 2019 still ticking on legacy nodes we forgot to decommission. Proper embarrassing. Cheaper to let 'em run than migrate the three remaining sites.

Your RackNerd lot probably had the same thought. "Lee, shall we email the stragglers?" "Nah, let sleeping badgers lie."

If it works it works. But don't ask for a custom PHP module, yeah?

Honey badger don't care... about downtime
prague1983

Member

btw I use Arch

#6

1,827 days on shared vCPU? Respect

oops: 0000 [#1] SMP
mediaaustin

Member

Deliverability Nerd

#7
prague1983 said:
Check your sending IP against Barracuda and Spamhaus.

Update: clean on Barracuda, clean on Spamhaus. One "poor reputation" flag on SenderScore but that's the whole /24, not my IP specifically. Gmail inbox rate for my newsletter last month was 94%. Not great, not poisoned.

The abuse desk thing is real though. Six months ago I reported a neighbor phishing via ticket. No response, no takedown. Still live at the same IP last I checked.

SPF, DKIM, DMARC — holy trinity ✉️
prague1983

Member

btw I use Arch

#8
web10 said:
A single vCPU at 100% for thirty seconds would probably trip whatever ancient OpenVZ container you're stuffed into.

It's KVM actually. I checked /proc/cpuinfo out of curiosity after the thread started. Intel Xeon E5-2680 v2, which is vintage 2013 but not OpenVZ.

Load average 0.08, 0.12, 0.09. Three static sites, one WordPress with LiteSpeed cache, one Gitea mirror. The Gitea pull cron is the most intensive thing I run.

oops: 0000 [#1] SMP
lee_mcr

Member

3 #9
jane_ffm said:
I'm half-convinced I'm their last paying customer and nobody's told me to leave.

There's your answer then. No abuse desk, no takedown, nobody home. You're not a valued customer, you're a rounding error in a spreadsheet from 2019.

I've been there. We had a cPanel server in 2021 where the billing integration broke and we didn't notice for eight months. Thirty-odd "customers" getting free service, none of them saying a word. Eventually the hardware failed and we just... didn't rebuild it. Emailed the ones with working addresses, most bounced.

You're one hardware failure away from evaporating.

Honey badger don't care... about downtime
jane_ffm

Member OP

#10
lee_mcr said:
You're one hardware failure away from evaporating.

I know. I've got rsnapshot pushing to Contabo's object storage in Nuremberg nightly, plus a weekly rclone to Hetzner. The sites would be back in 24 hours on any fresh VPS.

That's the thing though — it's been "one hardware failure away" for 1,827 days. At some point the probability estimate needs updating.

Single mode till I die 💀

Post a reply

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

Post reply Preview Save draft