ismailarif
DevOps and infrastructure engineer. Linux, Kubernetes, networking and storage, on-prem and hybrid.
# about
I’m a DevOps and infrastructure engineer working across on-prem and hybrid environments. Day to day that means Linux, a bare-metal Kubernetes platform, a hyper-converged Proxmox VE and Ceph cluster, and the networking that ties it all together.
I also support the applications running on top: Frappe/ERPNext and other Python, JavaScript and TypeScript apps. When a site is slow or a pod won’t start, I work with developers on debugging, database tuning and Git workflows to get it fixed.
On the security side I run Wazuh for monitoring and step-ca for internal PKI, and keep my own laptop hardened with TPM-backed encryption, Secure Boot and USB device control.
This site is where I write up that work: what broke, what I changed and what I learned.
skills
- linux & systems
- Ubuntu, Debian, systemd tuning, kernel optimisation
- kubernetes
- bare-metal clusters (kubespray / kubeadm), MetalLB, Istio, Gateway API, external-dns, cert-manager, dual-stack networking
- virtualization
- Proxmox VE, Proxmox SDN, Proxmox Backup Server, VMware ESXi (migration), govc, Dell iDRAC / racadm
- storage
- Ceph, Rook-Ceph, NFS, RAID, S3 API (Ceph RGW)
- databases
- MariaDB, PostgreSQL, query and server tuning, slow-query analysis
- backups
- Proxmox Backup Server, pgBackRest, S3-compatible object storage
- networking
- IPv6, BGP, IPsec / IKEv2, SD-WAN, firewall / routing, OPNsense, tcpdump, DNS troubleshooting
- app support
- Nginx, Gunicorn, Node.js, Frappe / ERPNext (v12, v16), Frappe app development, Documenso, report and print performance, Docker image builds
- automation & ci/cd
- Bash, Python, API integrations (Proxmox, FortiOS), GitHub Actions, ArgoCD
- security
- Wazuh SIEM, TLS / PKI (step-ca), incident response, Fortinet WAF, LUKS + TPM, Secure Boot, usbguard
- git
- rebase, reset, conflict resolution, filter-repo, developer workflow guidance
# projects
Bare-metal Kubernetes platform
A bare-metal Kubernetes platform with dual-stack networking, Istio, and DNS names and TLS certificates issued automatically.
Hyper-converged Proxmox and Ceph
Turning a 3-node Proxmox VE cluster into hyper-converged compute and storage: from 30k IOPS on a direct SSD to 80k on Ceph.
ESXi to Proxmox VE migration
Moving roughly 50 production VMs off VMware ESXi onto a Proxmox VE cluster backed by Ceph.
Rook-Ceph storage for Kubernetes
Persistent block and shared volumes for Kubernetes, served by the Proxmox Ceph cluster through Rook in external mode.
ERPNext and MariaDB performance
Fixing report and request timeouts across a fleet of Frappe/ERPNext sites.
Frappe internal tools
Internal Frappe apps and dashboards, built and shipped as Docker images.
# experience
- 2023-present DevOps Engineer · Tektician Sdn. Bhd.
education
- 2019–2024 Bachelor of Computer Science (Hons.) Data Communication & Networking