One thing you realize when you work by yourself is that you get a lot of benefits from tools that can automate some of your work or save you time. I have grown quite familiar with tools that you want around you when developing great PHP applications. With your production servers, you want services and products that can help you with:
- Server monitoring
- Server provisioning
- Log management
- Checking for uptime
- Application performance
- Application errors
This doesn’t even include developer specific services like continuous integration.
Those production server errors
Here is a common scenario.
Our client just called saying they had an error in production.
I can’t count the number of times I heard that quote. Production server errors in PHP (often the white screen of death) have always been a huge problem. When you get production errors in PHP, it’s a very painful experience for a lot of reasons.
Often, you weren’t aware of them so you have to go comb through your logs to see if you can find anything useful if you have log management that step is easier. If you’re a bit more on top of it, you get an email with a stack trace. But after, you have to go through this awkward reverse engineering process where you’re missing a lot of key information. How do you get that information?
For a lot of PHP devs (especially in the WordPress world), you will doing var dumps directly in production. That’s cowboy coding and that’s bad! Often, it’s necessary because you just don’t know what’s going on. Log information and a stack trace is sparse at best.
Meanwhile on the web…
I asked the Montréal WordPress Facebook Group what they used for errors, but only one person used an error monitoring service (Airbrake). I am always checking what the other big web languages (mainly Ruby and Python) are using to solve their problems and there are a lot of general tools that were built for this (e.g. Raygun and Sentry) that come with their own PHP libraries or third-party ones. In my research, I came across one tool, Honeybadger, that did what I wanted and they had one awesome policy.
We believe that an error management system built specifically for Ruby will be 10x better than any one-size-fits-all option.
I thought I want something like that for PHP and its frameworks/platforms. Where it’s someone who actually uses PHP and, most importantly, wants to use it. There’s plenty of language options out there and even more PHP haters so it would be nice to have someone who’s building a product where PHP isn’t an afterthought.
Hey, I know this problem! …and I know PHP!
And all I want is for something like Honeybadger to exist in PHP. So when I was brainstorming my B2B business for the app challenge, this was the first thing I thought about.
So that desire to see this exist got me to create Helthe.
Helthe helps you fix those production errors
My goal is to make Helthe the world-class error monitoring service that PHP deserves. You’ll get more than an alert, a log entry or a stack trace. You’ll get a tool designed to give you the information you need to solve those bugs faster without all the hassle.
If you are interested in having a better handle on your production server issues, check out Helthe.co for details. Otherwise, just keep checking back on how things are going. I’ll be posting regularly and sharing as much as possible of what I have been learning.
I will also be posting about production server topics and other tools you can use. So stay tuned!
Feel free to leave a comment below or discuss on Hacker News.
P.S. If you’d like to keep up to date with my progress, leave your email below!