521, 522 or 525 errors
Cloudflare cannot reach your origin, the connection times out, or the SSL handshake fails — and visitors see a Cloudflare error page instead of your site.
[ Servers & DevOps ]
Orange cloud on. Everything else configured too.
Turning on Cloudflare takes five minutes. Getting it right takes longer — SSL mode set to Flexible, a WAF running on defaults, a cache hit ratio in single digits and an origin IP still sitting in public DNS history. I set it up the way it is meant to work, and fix it when it is already half-working.
Free plan or Enterprise · No downtime on the switch · $20/hr or fixed price
[ Why you're here ]
Most of these come from a setup wizard clicked through quickly, or rules added one at a time over years.
Cloudflare cannot reach your origin, the connection times out, or the SSL handshake fails — and visitors see a Cloudflare error page instead of your site.
ERR_TOO_MANY_REDIRECTS after enabling Cloudflare, almost always Flexible SSL fighting an HTTPS redirect on the server.
Credential stuffing on /wp-login.php or /admin, scrapers hammering search pages, or a layer 7 flood the default settings let straight through.
HTML never cached, a cache hit ratio under 20%, and every request still traveling to an origin on the other side of the world.
wp-admin, phpMyAdmin, staging sites and dashboards reachable by anyone who guesses the URL.
Legacy Page Rules, overlapping firewall rules and a Worker someone deployed once, with no record of why any of it exists.
[ How it works ]
Your current DNS, origin, SSL setup and traffic reviewed. If you are already on Cloudflare, every existing rule is read and either kept, rewritten or removed — with the reason written down.
DNS moved or cleaned up, origin certificate installed, WAF and rate limiting rules written for your actual application, and caching rules tuned so pages are served from the edge.
Error pages, redirects, forms, logins, payments and webhooks tested through Cloudflare. You get a short document explaining every rule and setting, so the next person is not guessing.
[ What actually changes ]
Scope depends on your plan and your stack. This is what a typical setup or clean-up covers.
[ Example engagement ]
A WooCommerce store on Cloudflare with Flexible SSL, intermittent 522 errors at peak and bots hitting the login page. Moved to Full (strict), allowlisted Cloudflare IPs on the origin firewall, added rate limiting and rebuilt the caching rules.
Illustrative example of a typical engagement. Figures vary with the state of your systems and are not a guarantee of a specific outcome.
[ Free · quick look ]
Send me the domain and the error code. I will tell you what is most likely causing it and how long the fix should take, before you commit to anything.
[ Pricing ]
Tell me the number you have to work with. I'll tell you honestly what's achievable within it — and if it isn't enough, I'll say so before we start rather than halfway through.
Hourly rate $20/hoursmall jobs and ongoing work · fixed price for defined projects
Scope agreed in writing, price agreed in writing, before any work starts. No hourly creep and no invoice you haven't already approved.
For ongoing work — maintenance, monitoring, updates and small changes. Month to month, cancel whenever, no minimum term.
For a single bug, a short task or ongoing tweaks, where writing a full scope would cost more than simply doing the work. Billed in real time, reported in writing.
Budget too tight for the whole thing? I'll often suggest doing the highest-value part first and the rest later, rather than doing all of it badly.
[ Questions ]
Small jobs and fixes, like a 522 error or a redirect loop, are billed at $20 per hour and most take one to three hours. A full setup or a clean-up of an existing account is quoted as a fixed price, agreed in writing before I start, so you know the total up front.
521 means your origin refused the connection — often a firewall blocking Cloudflare IPs, or the web server being down. 522 is a timeout, usually an overloaded origin or packets being dropped. 525 is an SSL handshake failure between Cloudflare and your server, typically a missing or mismatched certificate under Full (strict). Each has a short list of causes and I work through them in order.
It should not. I copy and verify every record before changing nameservers, leave the old provider's zone intact until propagation is complete, and check email records in particular, because MX and SPF mistakes are the most common thing that breaks.
Often not. The free plan covers DNS, SSL, custom WAF rules, caching and Zero Trust for small teams. Pro adds the managed WAF ruleset, and APO is a low-cost add-on. I will tell you whether a paid plan is worth it for your traffic before recommending one.
Almost always because SSL is set to Flexible: Cloudflare talks to your server over HTTP, your server redirects to HTTPS, and the loop begins. The fix is a certificate on the origin and a switch to Full (strict), not turning off the redirect.
Usually, yes. With the site proxied, the origin locked down and a few targeted rules or Under Attack mode, most layer 7 floods drop off quickly. If your origin IP has already leaked, it may also need to change, and I will tell you if that is the case.
It can if caching or WAF rules are careless. I bypass cache for carts, accounts and APIs, allow payment provider and webhook traffic through where needed, and test those flows through Cloudflare before handover.
Yes. I work remotely with clients in the US, UK, Germany, Canada, Australia and elsewhere. Cloudflare changes are made in agreed windows that suit your time zone and traffic pattern.
[ Related services ]
[ Guides on this topic ]
[ Let's talk ]
Send the domain and the problem in a few lines, and you'll get a real reply from the person who'd do the work — same working day, no discovery call required.