Home
Sepehr’s DevOps field notes
Sepno NetworkEngineering calm into complex systems.
I write about cloud infrastructure, delivery pipelines, Kubernetes, observability, and the lessons hidden inside production systems.
Kubernetes / healthydeploy / automatedOperating principles
Reliability is designed, not wished for.
DevOps works when delivery, infrastructure, security, and feedback become one connected system.
Automate the path
Build short, repeatable routes from commit to production with useful checks at every step.
Observe the truth
Connect metrics, logs, traces, and service objectives to decisions engineers can act on.
Design for recovery
Treat failure as system feedback, then make the next incident smaller and easier to understand.
Latest writing
Notes from the build.
Architecture breakdowns, practical guides, tooling choices, and honest lessons from operating software.
-
Tailscale VPN: Secure Docker Swarm Networking Without Public IPs
A practical guide to connecting Docker Swarm nodes across different networks with Tailscale—without public IPs, router port forwarding, or exposed management ports.
-
اتصال امن Docker Swarm با Tailscale VPN؛ بدون IP عمومی و Port Forwarding
تجربه عملی اتصال امن نودهای Docker Swarm در شبکههای مختلف با Tailscale؛ بدون IP عمومی، Port Forwarding یا بازکردن پورتهای مدیریتی روی اینترنت.
The working toolbox
From commit to cloud.
Tools change. Clear interfaces, fast feedback, and operational ownership keep paying dividends.
LinuxDockerKubernetesTerraformGitHub ActionsPrometheusGrafanaAnsibleAWSAzureGitOpsDevSecOps
A practical loop
Build the system that builds the system.
Make change small
Keep work reviewable, testable, and safe to release without ceremony.
Make signals useful
Measure what users experience and what operators need to decide next.
Make learning durable
Turn incidents and experiments into better defaults, documentation, and automation.
“You build it, you run it” is not a burden. It is the shortest feedback loop in engineering.Sepno Network / DevOps Knowledge