Starting server fails on mac os x when starting pods

[titusn]

I tried using chectl to start a stand-alone che server. I previously managed to get it running just using the old-fashioned docker run command. That worked, but I ran into trouble creating projects (see https://stackoverflow.com/questions/56415583/eclipse-che-7-node-js-error-this-workspace-is-using-old-definition-format-which), which seemed fair, as I was following a method meant for che 6.

During start up I consistently got the error ERR_TIMEOUT: Timeout set to pod ready timeout 130000 when chectl tried to start the pods. I tried:

• deleting all docker containers, cache, volumes and images
• reinstalling minikube
• running minikube by hand (see error log below)
• deleting the chectl server
• raising memory of minikube to 4096 MB
• raising the pod timeout
• using a different kubernetes cluster (unfortunately docker4mac was not supported, even though the quick-start guide suggests it should be possible)

Expected behaviour

chectl is able to start the pods after downloading the images

Observed behaviour

chectl errors out:

chectl server:start --k8spodreadytimeout=500000

✔ ✈️  Minikube preflight checklist
✔ Verify if kubectl is installed
✔ Verify if minikube is installed
✔ Verify if minikube is running
↓ Start minikube [skipped]
  → Minikube is already running.
✔ Verify if minikube ingress addon is enabled
↓ Enable minikube ingress addon [skipped]
  → Ingress addon is already enabled.
✔ Retrieving minikube IP and domain for ingress URLs...192.168.99.100.nip.io.
✔ 🏃‍  Running Helm to install Che
  ✔ Verify if helm is installed
  ✔ Create Tiller Role Binding...it already exist.
  ✔ Create Tiller Service Account...it already exist.
  ✔ Create Tiller RBAC
  ✔ Create Tiller Service...it already exist.
  ✔ Preparing Che Helm Chart...done.
  ✔ Updating Helm Chart dependencies...done.
  ✔ Deploying Che Helm Chart...done.
❯ ✅  Post installation checklist
  ❯ Che pod bootstrap
    ✔ scheduling...done.
    ✔ downloading images...done.
    ✖ starting
      → ERR_TIMEOUT: Timeout set to pod ready timeout 500000
    Retrieving Che Server URL
    Che status check
Error: ERR_TIMEOUT: Timeout set to pod ready timeout 500000
    at KubeHelper.<anonymous> (/snapshot/chectl/lib/api/kube.js:0:0)
    at Generator.next (<anonymous>)
    at fulfilled (/snapshot/chectl/node_modules/tslib/tslib.js:104:62)

Steps to reproduce

1. Under Mac OS High Sierra use homebrew to install minikube, helm and chectl.
2. Run chectl server:start

[ye1984]

Same issue.... and same config as you

[azatsarynnyy]

There's Che Server issue which is already fixed but a docker image hasn't been updated yet.
So if you don't require to get the latest changes I can suggest you start Che with 7.0.0-beta-5.0 version for example.
chectl server:start --cheimage=eclipse/che-server:7.0.0-beta-5.0

[titusn]

Thank you for the answer. I can confirm that starting that version works. The first time I tried I did get the following error:

❯ ✅  Post installation checklist
❯ Che pod bootstrap
  ✔ scheduling...done.
  ✔ downloading images...done.
  ✖ starting
    → Get pods by selector "app=che" returned a pod with an invalid status.conditions
  Retrieving Che Server URL
  Che status check

Error: Get pods by selector "app=che" returned a pod with an invalid status.conditions
at KubeHelper. (/snapshot/chectl/lib/api/kube.js:0:0)
at Generator.next ()
at fulfilled (/snapshot/chectl/node_modules/tslib/tslib.js:104:62)

On a retry it just worked. The first time it seemed like downloading the image took almost no time, so probably it didn't download the other image immediately.

[azatsarynnyy]

FYI: nightly image has been already updated, so chectl server:start w/o params successfully starts Che Server

[ye1984]

@titusn same as you :) it works with the image given by @azatsarynnyy