Skip to content
Register

I became a sysadmin because of this forum

General Discussion by glasgow_tom 24 replies 3.8K views
glasgow_tom

Member OP

#1

Source engine server admin in 2012, 128 tickrate on a $2 vps from some dead provider. Tickrate explanations unprompted: sv_minrate 80000, sv_maxrate 0, that's unlimited. Kids play valorant now, no community servers, no forums like this. Learned linux because the css server wouldn't start without it. Now I'm a six-figure sysadmin and it started with lying about knowing bash on my resume. The lie became true because I broke everything and fixed it. Ask me anything I guess, though most of you won't like the answer.

uma

Member

99.99% or bust

#2

What was your uptime percentage on that first vps? Mine was 99.17% and I thought that was acceptable. Now I get paged if it's below 99.99%. Status page green became an addiction. Did you monitor from the start or was it all reactive? Alert fatigue hit me around year three, curious if you planned for it or just endured.

436 days. reboot is surrender.
danfra

Member

#3
glasgow_tom said:
Now I'm a six-figure sysadmin

2 gbps scrubbing at the provider edge, no upstream filtering. Learned that the hard way when a game server I ran got hit. Moved to anycast after. Your six-figure job, do they run their own scrubbing centers or leverage cloud? 500 gbps attacks are baseline now. Layer 7 is where most damage happens anyway.

mitigated 800Gbps before breakfast
ana_mad

Member

1 #4

Optimization matters. Electricity costs were my first real infrastructure concern, back when. Learned power delivery and cooling before I learned networking, honestly. The workload isn't important but the efficiency is. Your path sounds familiar.

swimming upstream since 2019 🐟
BI

Member

3 #5

Y'know? Like I have a $3 vps from CloudCone and I just run nginx and break it tbh? Imo that's learning but like is it? My resume is empty except for a python script that doesn't work? Should I just lie too or like is that bad? Everything feels impossible?

fanboy

Member

#6

Watt per core, that's the metric. Your six-figure job, do they run arm yet? Apple Silicon proved it, x86 is dead weight. Intel's burning electricity for nostalgia.

  • I run my whole lab on rockchip boards, 5w per node.
  • Dismiss compatibility concerns all you want, the future is arm
  • You're behind if you're not planning migration.
Angry at your employer probably.

ana_mad

Member

#7

Lol this is wild, you just lied? I lie on my LinkedIn all the time, say I know Kubernetes when I just ran minikube once. Broke my first VPS in 2 hours, rm -rf / but like on purpose? I don't know why I do this. Now I got a job fixing things I broke on purpose, full circle. Your path gives me hope tbh

swimming upstream since 2019 🐟
Joy

Member

#8

99.17% and you thought that was acceptable? I'd have cried

ana_mad

Member

#9

Lied about bash and the lie became true — that's the whole thread right there

swimming upstream since 2019 🐟
glasgow_tom

Member OP

#10
uma said:
Did you monitor from the start or was it all reactive?

Reactive. I had a cron job that emailed me when disk hit 90%, that was it. The real monitoring came after I slept through a 4am outage and 200 teenagers flamed me on Steam. Now it's Prometheus and Grafana like everyone else, but I still check the dashboards too often.

Post a reply

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

Post reply Preview Save draft