Can't get Eclipse Che to install

Question

I'm trying to install Eclipse Che on an existing Kubernetes cluster with two nodes using chectl. I have some of the applications that Eclipse Che requires already installed (and working) on my cluster; Postgres, Keycloak (as OIDC), and Cert Manager. I have provided a patch yaml to chectl as follows (secret, password, and auth URL changed);

apiVersion: org.eclipse.che/v2
spec:
  networking:
    domain: kubernetes.default.svc.cluster.local
    annotations:
        kubernetes.io/ingress.class: nginx
    auth:
      identityProviderURL: https://sso.mydomain.com
      oAuthClientName: eclipse-che
      oAuthSecret: BBz8ZfGzGXY0vUfeoKgP8vczFrj3D5KU
    database:
      externalDb: true
      chePostgresHostname: 'postgres.postgres.svc.cluster.local'
      chePostgresPort: '5432'
      chePostgresUser: 'eclipse'
      chePostgresPassword: '83xf5tC85LA9ZX4w'
      chePostgresDb: 'eclipse'

I'm calling chectl with;

chectl server:deploy --platform=k8s --che-operator-cr-patch-yaml ~/che/patch.yaml --domain https://kubernetes.default.svc.cluster.local --skip-cert-manager --k8spodreadytimeout=500000 --k8spoderrorrechecktimeout=500000 --installer=operator --debug

The result is always the same (note this is not the first attempt, so some resources already exist);

› Current Kubernetes context: 'kubernetes-admin@kubernetes'
  ✔ Verify Kubernetes API...[1.26]
  ✔ Kubernetes preflight checklist
    ✔ Verify if kubectl is installed...[OK]
    ✔ Verify domain is set...[OK]
  ✔ Create Namespace eclipse-che...[Exists]
  ↓ Install Cert Manager v1.8.2 [skipped]
  ✔ Start following Eclipse Che installation logs...[OK]
  ❯ Deploy Eclipse Che operator
    ↓ Install Dev Workspace operator [skipped]
      → Dev Workspace operator already installed
    ✔ Create ServiceAccount che-operator...[Exists]
    ✔ Create RBAC
      ✔ Create Role che-operator-leader-election...[Exists]
      ✔ Create Role che-operator...[Exists]
      ✔ Create RoleBinding che-operator-leader-election...[Exists]
      ✔ Create RoleBinding che-operator...[Exists]
      ✔ Create RoleBinding eclipse-che-che-operator...[Exists]
      ✔ Create RoleBinding eclipse-che-che-operator...[Exists]
    ✔ Create Certificate che-operator-serving-cert...[Exists]
    ✔ Create Issuer che-operator-selfsigned-issuer...[Exists]
    ✔ Create Service che-operator-service...[Exists]
    ✔ Create CRD checlusters.org.eclipse.che...[Exists]
    ✔ Waiting...[OK]
    ✔ Create Deployment che-operator...[Exists]
    ❯ Eclipse Che Operator pod bootstrap
      ✔ Scheduling...[OK]
      ✔ Downloading images...[OK]
      ✖ Starting
        → Failed to start a pod, reason: Error, exitCode: 1
      Create ValidatingWebhookConfiguration org.eclipse.che
      Create MutatingWebhookConfiguration org.eclipse.che
      Create CheCluster Custom Resource
    Error: Command server:deploy failed with the error: Failed to start a pod, reason: Error, exitCode: 1 See details: /home/myuser/.cache/chectl/error.log. Eclipse Che logs: 
    /tmp/chectl-logs/1681617965184.

The outout of /home/myuser/.cache/chectl/error.log is;

2023-04-16T04:06:30.649Z Error: Command server:deploy failed with the error: Failed to start a pod, reason: Error, exitCode: 1 See details: /home/myuser/.cache/chectl/error.log. Eclipse Che logs: /tmp/chectl-logs/1681617965184.
2023-04-16T04:06:30.649Z     at Object.newError (/usr/local/lib/chectl/lib/utils/utls.js:41:19)
2023-04-16T04:06:30.649Z     at Object.wrapCommandError (/usr/local/lib/chectl/lib/utils/command-utils.js:53:19)
2023-04-16T04:06:30.649Z     at Deploy.<anonymous> (/usr/local/lib/chectl/lib/commands/server/deploy.js:122:44)
2023-04-16T04:06:30.649Z     at Generator.throw (<anonymous>)
2023-04-16T04:06:30.649Z     at rejected (/usr/local/lib/chectl/node_modules/tslib/tslib.js:165:69)
2023-04-16T04:06:30.649Z     at runMicrotasks (<anonymous>)
2023-04-16T04:06:30.649Z Cause: Error: Failed to start a pod, reason: Error, exitCode: 1
2023-04-16T04:06:30.649Z     at /usr/local/lib/chectl/lib/tasks/pod-tasks.js:192:35
2023-04-16T04:06:30.649Z     at Generator.next (<anonymous>)
2023-04-16T04:06:30.649Z     at fulfilled (/usr/local/lib/chectl/node_modules/tslib/tslib.js:164:62)
2023-04-16T04:06:30.649Z     at runMicrotasks (<anonymous>)

The output of the pod logs from /tmp/chectl-logs/1681617965184/eclipse-che/che-operator-69bfb7c98-thm9l/che-operator.log is (as a side note; adding --debug doesn't seem to change verbosity of the logging at all);

2023-04-16T04:01:13.725Z    ERROR   Unable determine installation platform  {"error": "could not read API groups: Get \"https://10.96.0.1:443/api?timeout=32s\": dial tcp 10.96.0.1:443: i/o timeout"}
runtime.doInit
    /usr/lib/golang/src/runtime/proc.go:6230
runtime.main
    /usr/lib/golang/src/runtime/proc.go:233

There are two pods created by chectl, but they always go into CrashLoopBackOff;

devworkspace-controller   devworkspace-controller-manager-685cb85df7-27cqc   1/2     CrashLoopBackOff   167 (4m49s ago)   12h
eclipse-che               che-operator-69bfb7c98-dg9wc                       0/1     CrashLoopBackOff   9 (3m3s ago)      29m

There doesn't seem to be any way to get transparency to what chectl is doing in order to manually step though and troubleshoot. Googling suggests that loads of people are having problems at the same point in the install, but there are no answers. I can only assume that they ultimately end up giving up and going for a different solution.

I'm at a loss as where to look next or how to troubleshoot this any further. Has anyone been able to get past this point in an Eclipse Che install?

Update: I've now also tried installing with Helm and got the same result. — ScorpioTiger, Apr 16 '23 at 10:17
I wrote up a whole long piece on the issues I faced and the solutions to them, but stackoverflow won't let me post it. I guess it's destined to be painful for anyone who attempts to get Eclipse Che up and running. — ScorpioTiger, Apr 23 '23 at 02:36

Francewhoa · Answer 1 · 2023-07-17T09:50:35.247

One of the easiest and fastest way to install Eclipse Che on Kubernetes:

chectl server:start -m -n che --platform=k8s

Below is the same as above. But with details if you're interested in those.

Steps:

Install Kubernetes infrastructures version 1.21 or more recent. Using this documentation at https://kubernetes.io/docs/setup/
Set up Users' Authentication Kubernetes infrastructures. Using the steps at https://kubernetes.io/docs/reference/access-authn-authz/authentication/
Install Eclipse Che Command Line Interface CLI (chectl). Using the steps at https://www.eclipse.org/che/docs/stable/administration-guide/installing-the-chectl-management-tool/
Using Terminal/Console, execute this one command line to install Eclipse Che on Kubernetes (k8s)

chectl server:start -m -n che --platform=k8s
- Note: Depending on both your present needs and operating system, if appropriate, adapt this command line above. Using this documentation at https://github.com/che-incubator/chectl/#table-of-contents

Can't get Eclipse Che to install

1 Answers1

Steps:

Attributions: