Operations | Monitoring | ITSM | DevOps | Cloud

Your domain is on a spam blocklist. Password reset emails aren't arriving, order confirmations land in spam, and customers are complaining that "your site doesn't work." By the time you hear about it, the damage has been building for days. We've shipped DNS blocklist monitoring to catch this early. Oh Dear now checks your domain against 11 major blocklists and notifies you the moment you're listed, with direct links to get removed.

Our documentation, FAQ, and blog posts can now be served as clean markdown to AI agents. Send Accept: text/markdown or append.md to the URL, and you get structured content instead of a full HTML page. It worked great in development. Then we deployed, and two separate caching layers broke everything.

This is the final part of our 3-part series on SQL performance improvements. In part 1, we covered how to identify slow queries. In part 2, we explored how to fix them with indexes. In this post, we'll share how we prevent those performance issues from ever reaching production again. A few weeks ago, we massively improved the performance of the dashboard & website by optimizing our SQL queries.

When you monitor a website at Oh Dear, the monitoring also includes the historical performance insights that belong to that monitor. It gives you a historical overview of the speed of that monitor, allowing you to see anomalies and changes over time. As of today, there's a second view available, one that matches the webbrowser visualisation of the timing of a single request. This view shows the same waterfall information you'd find in Chrome or Firefox, providing a familiar view to developers worldwide.

Would you know if your checkout form stopped working overnight? Or if a recent deploy broke your login flow? Traditional monitoring can't catch these issues - it only tells you if your site is up, not if it actually works. AI monitoring lets you describe what should work in plain English, and we'll test it like a real user would - clicking buttons, filling forms, checking content. No scripts to maintain, no complex setup.

This is part 2 of a 3-part series on SQL performance improvements. A few weeks ago, we massively improved the performance of the dashboard & website by optimizing some of our SQL queries. In this post, we'll dive deeper into the optimisations of queries with indexes.

We're obviously a little biased, but we believe we have one of the best website monitoring tools on the market today, leading in features compared to our competitors. We've already tried a variety of marketing techniques to promote our service, but none really had the impact we were looking for. Maybe we're better at actually building good software than we are at marketing it? Or are we trying what everyone else is also doing, thus making it all harder?

We have launched Request Mirror, a little free service to reflect HTTP requests. We've also open-sourced it: you can read the code in the ohdearapp/request-mirror.ohdear.app repo on GitHub. In this blog post I'd like to explain why we built it and how you can use it.

A few weeks ago, we massively improved the performance of the dashboard & website by optimizing some of our SQL queries. In this post, we'll share how we identified the queries that needed work. In the next post, we'll explore how we fixed each of them. We'll cover the basics and gradually work our way up to the more advanced/complex ways of identifying slow queries. In this post, you'll see: Let's go!

Today, we have launched our MCP server. MCP (Model Context Protocol) is a standardized way for AI models to connect with external data sources and tools. If you use a tool like Claude Code, then this is how you can connect Oh Dear to it (you can create an API token in your account settings)