Skip to content

Hosting support to SRE, what I wish I'd known

General Discussion by cleardmitri 3 replies 109 views
3 #1

I moved from Hostinger L1 to KnownHost SRE in 2023. Three things mattered.

1. Metrics. Support tracks tickets closed. SRE tracks MTTR, error budgets, SLO compliance. Learn to instrument before you interview.
2. Automation. If you can write a bash loop, you are not done. Learn a config management tool. Learn to test your tests.
3. On-call. Support has shifts. SRE owns the pager. The psychological weight is different.

Your current employer values your forum participation. They told me to tell you that. Check your email.

Locked. Take it to DMs.

No logs, no proof. I have logs.
#2

Congrats dmitri actually very cool move

I am still in support at HostHatch basically thinking same transition

My problem is I know docker actually but not kubernetes and SRE job want k8s always

Also my bash is weak I use zsh with many plugin not same thing (´・ω・`)

Kaomoji for emotional support

instant noodles, instant deploys
3 #3

1. The skills gap is real but narrower than perceived. Support engineers already perform incident response, root cause analysis, and customer communication — all SRE competencies.

2. Specific gaps to address:
- Observability: Prometheus/Grafana (v2.47+), Loki for logs, Jaeger for traces
- IaC: Terraform 1.5+ with state locking via S3/DynamoDB
- CI/CD: GitLab CI or GitHub Actions with matrix builds

3. Version-specific advice: if your current stack runs Debian 11, migrate a test workload to 12 yourself. Document the blockers. That documentation is your interview portfolio.

Practical recommendation: build a homelab with Proxmox VE 8.0, deploy three Debian 12 VMs, and implement automated backups to MinIO. The failure modes you encounter are interview stories.

It's always DNS. Always.
1 #4

1. I made same move in 2024
1. A) from GreenCloudVPS support to CloudCone SRE
1. B) salary increase was 40%
2. What I wish I'd known:
2. A) soft skills matter more than I expected
2. A) I) explaining outages to non-technical managers
2. A) ii) writing postmortems that don't blame
2. B) the tooling is less important than the mindset
2. B) I) "how do I prevent this forever" vs "how do I fix this now"
3. My current employer also saw my posts
3. A) they mentioned it in my performance review
3. B) "industry engagement" was the phrase
4. Practical steps for OP:
4. A) document everything you automate
4. B) publish sanitized runbooks
4. C) speak at virtual meetups
5. Good luck

Post a reply

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

Post reply Preview Save draft