My main goal in ЦМРТ as a linux engineer is to achieve zero downtime for existing business critical applications and to provide infrastructure for smooth migration from internally developed 1C-based RIS* to QMS HIS* When I just got this job, average business critical application uptime was around 98.5%. Now it is 99.9998%. To achieve this, the following was done: Networking layer • Our old network infrastructure did not provide enough redundancy and was too depended on ISP reliability. When one of ISPs failed, it took too much time to reconfigure our network (3 datacenters and over 200 remote branches all over Russia). We registered our own ASN* and negotiated transit traffic agreements with our existing ISPs. Now we announce our IP prefixes over BGP, so failover from one ISP to another takes about 30 seconds (and happens automatically). • Another crucial part was outdated networking equipment. Most problematic cisco devices (main router, some intracompany switches) were replaced with newer models. Our new network relies on MikroTik instead. Virtualization layer • Updated virtualization infrastructure with latest vSphere/ESXi • Deployed vMware monitoring & automation solutions (vRealize) • Balanced applications and failover domains around our infrastructure for faster switch-over to backup appliances. Application layer • Migrated business critical applications from ancient unsupported distros (such as CentOS 6, Debian Wheezy, Ubuntu 14.04, etc.) to modern Ubuntu version. • Deployed application-level alerting via Zabbix and performance monitoring via NetData. Due to vast size of our company, necessity of every step we had taken was proven beforehand. I pinpointed exact cause of each problem and proposed an optimal cost-efficient solution.

Accomplishments during my work with bquadro include: • Implemented centralized client system monitoring with Zabbix • Standardized protocols on various IT processes, including service operation, incident management, quality control and more • Full adoption of JetBrains development environment with GitLab CI deployment by all employees • Migrated 3rd-party password & account management to a local Bitwarden server • Developed business strategies and road maps for new services, which were never attempted by the company before (mobile development, system integration, complex IT-audits and more) • Migrated several physical servers to Proxmox VM (KVM/QEMU) • Created convenient management system for isolated containerized (LXC) environments for each developer • Implemented centralized differential backup & restore system (Duplicity, Bash) • Migrated network infrastructure to MikroTik with multihomed OSPF failover and automated BGP censorship circumvention, SSTP dial-in VPN for employees with radius authentication via Microsoft Windows Server AD backend • Implemented 1C-Bitrix deployment to Docker containers via GitLab CI for developers • Created & utilized JMeter load test scenarios for web-servers • Deployed monitoring, logging & alerting system via Zabbix, Grafana & Prometheus • Multiple cases of forensic work to determine exact person (usually 3rd party), whose actions resulted in system malfunctioning My usual daily occupancy in bquadro included: • Estimating work-hours needed for fulfilling contract (presale) • Documenting best practices • Developing middleware for our business needs • Teaching my subordinate employees on how to do their work better • Creating primary development & service strategies for new projects to minimize possible future issues • Attending to various business negotiations

• Migrated several physical servers to Proxmox VM (KVM/QEMU) • Created convenient management system for isolated containerized (LXC) environments for each developer • Implemented centralized differential backup & restore system (Duplicity, Bash) • Migrated network infrastructure to MikroTik with multihomed OSPF failover and automated BGP censorship circumvention, SSTP dial-in VPN for employees with radius authentication via Microsoft Windows Server AD backend • Implemented 1C-Bitrix deployment to Docker containers via GitLab CI for developers • Created & utilized JMeter load test scenarios for web-servers • Deployed monitoring, logging & alerting system via Zabbix, Grafana & Prometheus • Multiple cases of forensic work to determine exact person (usually 3rd party), whose actions resulted in system malfunctioning • Support services for implemented solutions

• Enabled centralized management of HP ProLiant chassis security by implementing Samba4 Active Directory for authentication.

• Created a web-hosting resell system with client isolation using Docker containers featuring a convenient webui administration panel (PHP, MySQL, bash, Python)

• Various subcontract work for their clients (mainly telephony and server migration) • Several security incident investigations, pinpointing the attack sequences and identifying the attackers

• Smart contract programming for Ethereum blockchain (Solidity) • Developed a cryptocurrency trading bot for BitMex.com platform utilizing automated machine learning of technical analysis signals via hyper-parameter sci-kit forest regression on existing candle pandas series (Python) • Created 2.5 MW Bitmain Antminer rig inside a standard shipping container

• Interacting with various organization employees and focus groups to design optimal solution for a given enterprise in a given budget • Creating procurement management plans for required hardware • Documenting business processes, creating internal documentation & training programs for seamless integration • Providing various support for deployed solutions

• Development of modern operator script creation/usage system for Avaya/Oracle-based call-center (Java, Tomcat) • Developed a complete call-center infrastructure/software from scratch based on SIP.js frontend and asterisk/GoIP backend (PHP, MySQL) • Reverse engineer GoIP simbank firmware to enable root access to the system (binwalk, IDA Pro)

Информационных технологий, Информационные системы и технологии