cloudenza.NOTES / WEB REFERENCE
A SMALL, OPEN REFERENCE LIBRARY VOL. 01

The web is easier
when you know why.

Clear notes on the pieces behind every website: names, connections, responses and the caches in between. No mystery, no unnecessary jargon.

01

Plain explanationsUnderstand the mechanism, not just the acronym.

02

Practical examplesCommands you can run and inspect yourself.

03

No synthetic metricsChecks report only what the browser actually sees.

THE LIBRARY / 01—04

Useful foundations.

Four short reads for understanding what happens between a URL and a loaded page.

NOTE 001 NETWORKING
4 MIN READ

What DNS actually does

How a domain becomes an address, why resolvers cache answers, and what TTL really controls.

DNSTTLResolvers
NOTE 002 SECURITY
5 MIN READ

A short guide to HTTPS

Certificates, trust, encryption, and why a successful connection has several distinct steps.

HTTPSTLSCertificates
NOTE 003 PERFORMANCE
4 MIN READ

Caching without confusion

Freshness, validation and the difference between saving bandwidth and serving stale content.

Cache-ControlETagCDN
NOTE 004 PROTOCOLS
5 MIN READ

Reading an HTTP response

Status codes tell you the category of a result. Headers explain how to interpret it.

Status codesHeaderscurl
001 / NETWORKINGDOMAIN NAME SYSTEM↑ Back to library

What happens during a DNS lookup?

A domain name is a human-friendly label. DNS provides records that help software find services associated with that label.

When you open a website, the browser or operating system typically asks a recursive resolver for the relevant records. The resolver may already have a cached answer. Otherwise it follows the DNS delegation chain until it can obtain an authoritative answer. The browser can then try to connect to one of the returned addresses.

A records carry IPv4 addresses, AAAA records carry IPv6 addresses, and CNAME records point one DNS name to another. A DNS answer does not prove a web server is running: resolution and connectivity are different checks.

IMPORTANT DISTINCTION

TTL is a cache lifetime for DNS records, not a promise that an address will become reachable or change everywhere at the exact same moment.

Inspect a DNS A record
dig example.com A +short

Further reading: DNS fundamentals and RFC 1035.

002 / SECURITYENCRYPTED TRANSPORT↑ Back to library

Why HTTPS needs certificates

HTTPS is HTTP transported over a protected connection, normally using TLS. TLS aims to authenticate the peer and protect data in transit.

During connection setup, a browser and server negotiate how to communicate securely. The server generally presents a certificate chain. The browser checks factors including the hostname, validity period and whether the chain leads to a trusted authority.

The URL's hostname matters: a certificate for a different domain does not automatically validate a new subdomain. Encryption also does not mean the website itself is trustworthy; it means the connection has specific cryptographic protections.

Inspect a site's response and certificate validation
curl -Ivs https://example.com/

Further reading: MDN: TLS and TLS 1.3 specification.

003 / PERFORMANCEHTTP CACHING↑ Back to library

Fresh content, fewer requests

A cache can reuse a previously received response. Whether it may do so depends on the response's caching rules and context.

Cache-Control: max-age=600 permits a response to be considered fresh for up to ten minutes, subject to other directives and the cache's behavior. Once stale, a cache may be able to revalidate using an ETag and an If-None-Match request header.

A 304 Not Modified response allows reuse of the stored representation. no-cache requires successful validation before reuse; no-store instructs compliant caches not to store the response. These two directives are not synonyms.

Read caching headers
curl -I https://example.com/

Further reading: MDN: HTTP caching.

004 / PROTOCOLSHTTP RESPONSES↑ Back to library

The meaning behind a status code

A web response contains a status code, headers and sometimes a response body. The status code is a starting point, not a full diagnosis.

200OK — the request succeeded.
301Moved Permanently — the resource has a new permanent URL.
404Not Found — the requested resource was not found.
429Too Many Requests — the sender is rate-limited.
502Bad Gateway — a gateway received an invalid upstream response.
503Service Unavailable — the server cannot currently handle the request.

For diagnosis, inspect the requested URL, response headers, intermediate proxies and application logs. A 502 alone cannot identify whether the underlying cause is a failed origin, a timeout or something else.

Display the status and response headers
curl -sS -D - -o /dev/null https://example.com/

Further reading: MDN: HTTP status codes.

A SMALL PRACTICAL TOOL

Check this connection.

Send a new request from your browser to this website and measure the time until its response arrives. This is a single observation, not a network benchmark or an uptime guarantee.

Runs locally in your browser. No external tracking or telemetry.
ORIGIN REQUEST / LIVE TESTREADY
—ms
HTTP RESPONSENot checked
REQUEST TIME—

A request is sent to the same host serving this page.

ABOUT THIS PROJECT

Written to be useful,
not impressive.

Cloudenza Notes is an independent reference collection about everyday web infrastructure. Each note focuses on concepts that can be checked against public technical documentation.

There are no advertised hosting plans, invented customer counts or fabricated service-availability reports here. This site is a set of real articles and a small browser-based diagnostic.

Back to top ↑