Skip to content
Register

Finally decommissioned last of our Tier II footprint

Datacenter Talk by adam20 12 replies 2.2K views
adam20

Member OP

Ex-Abuse Desk

14 #1

The ticket said: "URGENT: customer reports complete data center offline, 500+ down, business critical."

It was our own. Last of the Tier II fleet, a sad row of Contabo cabinets in a building where the "redundant" UPS shared a breaker with the hallway vending machine. We migrated the final workload Monday. Tuesday I drove out to pull drives. Wednesday the landlord called asking why our rent was late.

Eight years. Started with three cPanel shared boxes and a dream that turned into a compliance nightmare. No more 99.5% SLA hand-waving. No more explaining to auditors why "physical security" meant a bike lock on a cage.

The gear? Already in landfill per our 2024 sustainability policy. Leaseweb offered to take the RAM for "spares" but policy says shred, so shred we did. I kept one drive caddy as a paperweight. It holds down abuse reports nicely.

reported. resolved. repeat.
tarekcai

Member

spaces are for websites

#2

Congrats on the exit
From tier ii hell
I remember our last one
InterServer Secaucus
"concurrent maintainability"
Meant two guys with screwdrivers
Never again
Tabs over spaces
Always

bravomartha

Member

#3

CONGRATULATIONS ADAM I REMEMBER WHEN WE HAD OUR LAST TIER II FOOTPRINT IT WAS BACK IN 2019 AT THIS PLACE IN TAMPA I THINK IT WAS CALLED SOMETHING LIKE COBALT SOMETHING OR MAYBE IT WAS HOST SOMETHING I CANT REMEMBER THE NAME EXACTLY BUT THE IMPORTANT THING IS I WAS EATING A BAGEL THAT DAY IT WAS EVERYTHING BAGEL WITH CREAM CHEESE AND MY CAT MR WHISKERS WAS SITTING ON THE KEYBOARD THE WHOLE TIME I WAS TRYING TO DO THE MIGRATION AND THE WEATHER WAS TERRIBLE IT WAS RAINING AND THUNDERING AND I THOUGHT WE WOULD LOSE POWER BUT WE DIDNT AND THEN WHEN I FINALLY GOT THE LAST VM MOVED THE CAT KNOCKED OVER MY COFFEE AND I STILL HAVE THE STAIN ON THAT SHIRT AND PEOPLE ALWAYS ASK WHY I KEEP IT AND I SAY BECAUSE IT REMINDS ME OF WHERE WE CAME FROM AND NOBODY EVER READS THESE WHOLE THINGS ANYWAY BUT I FEEL BETTER HAVING TYPED IT ALL OUT

pam3

Member

Latency Witch

#4

[1] Observed similar migration at previous employer. Decommissioned 4.2e1 cabinets from Tier II facility with observed availability of 9.972e2 (three nines, not four).

RFC 2119 "MUST" and "RECOMMENDED" do not apply to Tier II "concurrent maintainability" claims. [2]

Footnotes:
[1] CloudCone, 2.024e3
[2] Uptime Institute Tier Standard: Topology, 2.023e0 edition

100% packet loss at hop 3
Chen

Member

☁️ multi-cloud mess

#5

Tier ii with shared vending machine breaker. Classic

JA

Member

#6

Contabo still exists?

SH

Member

#7

I had a InterServer once too. Never again

JA

Member

#8

Contabo still exists and they still advertise "99.996% uptime" on a single power feed. The St. Louis DC is just their Nuremberg config with American power outlets.

adam20

Member OP

Ex-Abuse Desk

#9
Jab4 said:
Contabo still exists?

Barely. We were on their dedicated range, not the VPS circus everyone knows — https://contabo.com/en/vps/. Same building, different floor, equally creative interpretation of "redundant."

The landfill thing stings though. Leaseweb took our new footprint, Amsterdam and Singapore, proper Tier III. Cost per rack tripled. Compliance auditor actually smiled. First time.

reported. resolved. repeat.
Chen

Member

☁️ multi-cloud mess

3 #10
adam20 said:
Leaseweb took our new footprint

Singapore Leaseweb? I have a box there. Power usage efficiency is decent but their cross-connect pricing to Equinix SG1 made me cry. Ended up keeping AWS Singapore for the latency-sensitive stuff and Leaseweb for the bulk storage. Five providers, five invoices, one headache.

Post a reply

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

Post reply Preview Save draft