Troubleshoot & run k8s commands from Slack using AI
Doctor Droid simplifies the process of troubleshooting your Kubernetes (k8s) cluster. With just a message to the bot, you can quickly check pod statuses, restart pods, run custom commands, or diagnose problems you may find challenging to debug on your own.
tl;dr (Full story after setup instructions)
Hello everyone, it's been 2 years since we've building Doctor Droid. Our first user demo call started with telling them, what if you can type in your Slack "/droid check if service x is doing ok" and oh man! what a journey it has been since then. We have tried many different ways & product solutions to solve this problem before landing on what we have today, which we think is the most superior experience to helps teams to troubleshoot k8s (kubernetes) related issues directly from Slack using AI.
Firstly, if you're looking to try, here's how you can try it:
(a) Sign up --> https://drdroid.io/
(b) Add our bot to your Slack channel
(c) Add your k8s cluster
(d) Start chatting with it
Note: If your k8s clusters are not accessible from the internet, you can still make it work! Read more about how our proxy service works here: https://github.com/DrDroidLab/dr...
Read more about how our AI agent works and the related integrations here: https://docs.drdroid.io/
Full backstory:
Over the last couple of years, we've tried multiple approaches to help users troubleshoot complex issues faster. The biggest highlights are these:
* First we tried to actually ask users to add structured logs in their code and send us that. (https://drdroid.io/kenobi) It did work with a couple of companies.
* Next, we created an open source framework that people could use to write their own automation / remediation scripts in a no-code way. https://github.com/DrDroidLab/pl... -- this one got a couple of really large enterprises using too.
But overall there were two challenges in these approaches:
* Often engineers themselves didn't know where to add code changes or what scripts to write until the issue actually comes up
* Configuration and setup was too long
What we did know though: Engineers are exceptional at building hypothesis, testing them and re-doing this process until they are successful!
That's what we banked upon and started iterating with some early users on how we could solve this.
If you want to setup a time to chat in more detail with our team, please select a time from our website: https://drdroid.io/
