Restic backups done right: encryption, retention, browsing and recovery
2026-08-01
A practical Restic tutorial for encrypted, compressed and verifiable backups on Linux, including automation, retention and real restore tests.
Internet is my LAN.
I am Daniele V. , a full-stack developer, Linux system administrator and technical problem solver based in Venice, Italy. I build systems that collect, understand, protect and retrieve information—across the public web and inside private company infrastructure.
For more than a decade I have designed and maintained web-scraping systems, browser automations and data-processing pipelines. I am at my best when the problem is awkward, the data is inconsistent, the application is undocumented or a generic product is not enough.
My approach is deliberately hands-on: analyse the real problem, reduce it to clear components, build the smallest reliable solution, test it under realistic conditions and leave behind a system that can be understood and maintained.
---
DVfile — working title, serious ambition
Your company already owns the knowledge. DVfile makes it searchable without taking it away.
Companies accumulate years of PDFs, invoices, contracts, scans, spreadsheets, email exports and project files in ordinary folders. The documents still exist, but finding the right one becomes slow, inconsistent and dependent on whoever remembers its name or location.
DVfile is the working title of an on-premises file intelligence and cataloguing engine. It is being designed to index documents where they already live, enrich them with searchable metadata and let users narrow a search progressively by date, supplier, customer, invoice number, document type, project, location or custom tags.
The archive is not swallowed by a proprietary cloud and the original directory structure does not have to disappear. Files remain under the company's control. OCR, classification, metadata extraction and language-model-assisted analysis can run locally when confidentiality requires it.
The ambition is larger than “search inside PDFs”: DVfile aims to become a private discovery layer over the company's existing files—fast enough for daily work, transparent enough to trust and deployable inside the customer's own infrastructure.
Private pilot projects are being evaluated now.
---
Immution: lightweight protection for small Linux servers
Immution is an open-source, server-side security tool designed for small VPSs and web servers that cannot justify a heavy security stack.
It analyses Apache access logs, scores suspicious behaviour and can block clients that repeatedly probe sensitive paths, generate abnormal error rates or behave like automated attacks. It works without JavaScript challenges and without forcing legitimate visitors through an external security service.
Immution grew out of a real operational need: keeping compact Linux servers responsive during abusive traffic and application-layer denial-of-service attempts. It combines practical log analysis, filesystem state, process supervision and simple defensive decisions that remain understandable to the administrator.
Free open-source security software for small servers. Donations are welcome. Paid installation, tuning and support are available.
---
Services
Web scraping and browser automation — more than a decade in the field
I have spent more than ten years building systems that acquire and process data from the web. The work has evolved from cURL, HTML parsing and headless browsers to Playwright, Selenium, Puppeteer and hand-crafted automation through the browser DevTools Protocol.
A production crawler is not merely a selector and a loop. I design queues, retries, session management, duplicate detection, evidence capture, validation, rate controls, change detection, structured output and monitoring. I have worked on data acquisition for brand protection, hospitality, insurance, reviews, reporting and statistical analysis, including projects where millions of pages or frequent layout changes made maintainability more important than a one-off result.
Custom software and process automation
I design focused web applications, internal tools, APIs and automation pipelines in PHP, Python, Node.js, Bash and SQL. Typical work includes replacing repetitive procedures, integrating systems that were never designed to communicate and turning an unreliable prototype into an operational service.
Linux servers, logs and filesystem investigation
I configure, troubleshoot and harden LAMP and LEMP systems, investigate performance problems and recover services that have become fragile over time. I am comfortable working directly with Bash, `grep`, `awk`, `sed`, `find`, `journalctl`, Apache and Nginx logs, systemd units, permissions, ownership, mount points, inodes and Linux filesystem behaviour.
This includes incident reconstruction, abusive-traffic analysis, log correlation, disk-usage investigations, backup and restore design, filesystem migrations and the small diagnostic scripts that often reveal more than a large monitoring dashboard.
Data processing and document intelligence
I create pipelines that collect, normalise, classify and search large quantities of semi-structured information. This includes business documents, reviews, reports, public web data and operational logs, with local or cloud language models only where they provide a measurable advantage.
VoIP, audio and video systems
My background also includes VoIP and media technologies: SIP and Asterisk environments, audio codecs, video codecs, H.264 streaming, transcoding, bandwidth constraints and the practical interaction between networks, latency and perceived quality.
Security, backup and technical rescue
Some systems do not need a fashionable rewrite. They need someone willing to understand the existing architecture, identify the real failure points and improve it without disrupting the business. I provide application-security reviews, server hardening, backup design, disaster-recovery work, targeted repairs, migrations, monitoring and long-term technical support.
---
The kind of work I value
I am most useful when the problem crosses several technical areas or has already resisted a generic solution. I prefer direct communication, measurable outcomes, simple architectures and honest trade-offs.
I can work on a clearly defined assignment, investigate an unclear technical problem or help shape a product from an early concept into a maintainable service.
Email: Email Location: Venice, Italy — remote work across Europe and beyond