
kwatch
Monitor & detect crashes in Kubernetes cluster instantly
26 followers
Monitor & detect crashes in Kubernetes cluster instantly
26 followers
👀 kwatch is an open source tool that helps you monitor all changes in your Kubernetes(K8s) cluster, detects crashes in your running apps in realtime, and publishes notifications to your channels (Slack, Discord, etc.) instantly
This is the 2nd launch from kwatch. View more
kwatch
Launching today
kwatch finds the one cause behind a Kubernetes failure and sends one message: what broke, why, who changed it, what to do next. To Slack, Teams, email and 50+ channels. Source available (ELv2); Pro free during early access.





Free
Launch Team

AWS Activate CreditsCredits that fuel your startup’s next milestone
Promoted

Hi, I'm Abdelrahman, the maker of kwatch. We launched here five years ago as a way to hear about crashing pods. It did that, and every alert was true, but when a release broke something you still got fifty of them and had to work out why yourself.
So in v1.2.0 I rebuilt kwatch around the cause. It runs as one small pod in your cluster and keeps a live map of what depends on what. When something fails, it follows the failure upstream and sends one message: what broke, why, who changed it, and what to do next, ending with the command to run. When it is fixed you hear once, then it stays quiet.
Your events and logs are analyzed where they live. Alerts go to the channels you set up: Slack, Teams, Discord, PagerDuty, email and 50+ more.
To try it, run one command and it walks you through the rest:
bash <(curl -fsSL https://kwatch.dev/kwatch.sh)
kwatch is source available (ELv2). Community is free, and Pro is free during early access.
I would love to hear what your last 03:00 incident looked like, and whether the message would have helped.
where does the "who changed it" part come from, k8s events/audit log or does it hook into argo/flux too? most of our changes go thru gitops so the commit is what id actually want in the alert
@bacelj Good question. kwatch reads it from the cluster itself, not the audit log: the field manager that made the change, the change-cause annotation, and the GitOps application that owns the workload when Argo CD or Flux manages it. So for a GitOps change you see which app rolled it out, and the next step points you to a revert in Git rather than a kubectl rollback. It doesn't pull the commit itself from your Git host today; thanks for the nudge, that's useful to hear.