if you see warning: skipped value for secrets: Not a table. - don't worry, it is lying, it is related to brigade secrets and go unmarshaling, seems that k8s is not happy to have a hashmap instead of array, but go expect to have a hashmap - so all good!
getting error: a 'x86_64-linux' with features {} is required to build '/nix/store/vxwxcykyhdbiwyysj8fad14m0ynq6wlq-yarn.nix.drv', but I am a 'x86_64-darwin' with features {benchmark, big-parallel, nixos-test} (use '--show-trace' to show detailed location information)

you need to run remote worker since most likely you are on darwin

solution: run source run-linux-worker.sh

if your cluster is dying it seems that you have not sufficient resources for docker

solution: https://bitbucket.org/digitalrigbitbucketteam/embracing-nix-docker-k8s-helm-knative/issues/1/required-docker-resources-for-local

Error getting host status: state: docker-machine-driver-hyperkit needs to run with elevated permissions. > solution: Please run the following command, then try again: sudo chown root:wheel /nix/store/3l4b2dqvdrlbikmdd7xiawmbrwiqgz3j-minikube-1.0.1-bin/bin/docker-machine-driver-hyperkit && sudo chmod u+s /nix/store/3l4b2dqvdrlbikmdd7xiawmbrwiqgz3j-minikube-1.0.1-bin/bin/docker-machine-driver-hyperkit
building with kaniko and nix on mac os require adding /nix to shared path in docker
brigade.js got stuck and getting pending status when kubectl get pv -A

minikube addons enable storage-provisioner -p $PROJECT_NAME

aws skewed response -> orbs-network/orbs-network-go#1072 (restart docker?)
when terraform cannot delete subnets and ig - clean ELB first - don't know yet how to remove them automatically since they are requested from kubernetes context

issue with ssl and nix channels

ssl from go client in k8s

apiVersion: v1
clusters:
- cluster:
    server: https://192.168.0.100:8443
    insecure-skip-tls-verify: true
  name: minikube

cannot delete tunnel

minikube "x509: certificate signed by unknown authority"

Provide feedback

Saved searches