We have a client application that’s getting more complex and whose application is becoming a core part of their business’ continuity. As such, we’re looking to mature in how we manage our application reliability.
I’m curious if anyone else here has thought about this and what type of solutions they use. For context, here’s where we are now—
Up-time monitoring: We’re using BetterUptime to ensure our WeWeb projects are consistently live. This has proven invaluable, particularly during WeWeb’s downtime earlier this year, enabling us to proactively communicate with clients and work with the WeWeb team.
Error and Performance Monitoring: We’re in the process of identifying suitable solutions. Our back-ends are built on Xano, and we’re setting up an error and performance monitoring system using their metadata API’s request history endpoint. For the front-end, we’re considering tools like LogRocket or Sentry. While there are numerous other tools, most seem tailored for larger, code-centric applications.
A new player in the space is highlight.io. It’s on my list of things to try.
It has session replay, error monitoring and logging. It’s open source, you can self host, but in the docs they say the provided self hosting setup is not recommended for high volumes, where they obviously recommend thei cloud offering.
From a quick read of the docs it looks like it can be easily integrated in weweb’s frontend. For xano you probably need to send the error logs yourself using a http request.
Thank you, @dorilama. Highlight looks just what I’m looking for. For me, I think it comes down to deciding if I really need some of the product analytics features or not. I feel like I’d lean towards LogRocket if I want those, and towards Highlight if I don’t need them.
From a quick read of the docs it looks like it can be easily integrated in weweb’s frontend. For xano you probably need to send the error logs yourself using a http request.
I’m curious how you think about it, but I don’t like the idea of having to configure various error handling logic on every API endpoint.
And actually…hacking around with the metadata API, I was actually able to configure exactly what I needed yesterday (a simple dashboard to get an overview of errors and API performance) with Google Cloud Functions and BigQuery!
We’re just using the in-house back-end monitoring tool for now @ben.
We de-prioritized implementing a front-end tool, but I played around with https://livesession.io/ a bit, and I liked it as it was also helpful from a product analytics perspective.
I assume this topic is going to be hot now
So, encourage everybody to share their experience and actual results with their monitoring tools with recent downtime.
We personally had BetterStack and it completely missed the incident. It still says we are fine, lol. So, we are definitely going to change it with something else, but what?..
