Skip to content

Add warning message in init container if database has not been initialized

We currently do not show any log entry, in the event that we fail to initialize containers due to the database not being initialized.

We should add an appropriate log entry, so users can more easily debug why it is failing to start.

Original issue

Summary

I am trying to deploy gitlab community edition on bare-metal kubernetes cluster using the gitlab helm chart.Data is persisted locally with persistent volumes directly attached to local disks.All gitlab components are starting normally except the dependencies init container of unicorn and sidekiq

Steps to reproduce

Just deploy the chart with the configuration mentioned bellow after creating local persitent volumes.

Configuration used

global:
  edition: ce
  hosts:
    domain: code-test.example.com
    service:
      type: ClusterIP
  minio:
    enabled: true
  smtp:
    enabled: false
  appConfig:
    lfs:
      enabled: false

minio:
  persistence:
    storageClass: local-storage
    size: 200Gi
    matchLabels:
      component: minio

certmanager:
  install: false
  
certmanager-issuer:
  email: xx.yy@example.com

nginx-ingress:
  enabled: false
  
prometheus:
  install: false

postgresql:
  install: true
  persistence:
    existingClaim: gitlab-pgsql-local-pvc
  metrics:
    enabled: false

registry:
   enabled: false

redis:
  enabled: true
  persistence:
    storageClass: local-storage
    size: 200Gi
    matchLabels:
      component: redis

shared-secrets:
  enabled: true
  rbac:
    create: true

gitlab-runner:
  install: false

gitlab:
  migrations:
    enabled: false
  unicorn:
    enabled: true
    workerProcesses: 1
    maxUnavailable: 1
    minReplicas: 1
    maxReplicas: 2
    metrics:
      enabled: false
  sidekiq:
    enabled: true
    metrics:
      enabled: false
  gitaly:
    enabled: true
    persistence:
      storageClass: local-storage
      size: 200Gi
      matchLabels:
        component: gitaly
    resources:
      limits:
        cpu: 200m
        memory: 200Mi
      requests:
        cpu: 100m
        memory: 128Mi
  gitlab-shell:
    enabled: true
  task-runner:
    enabled: true

Current behavior

The dependencies unicorn init container fails to start after visualizing the log message specified in the log section.

Expected behavior

Everything should work correctly

Versions

  • Chart: ad7042f7
  • Platform:
    • Cloud: No Cloud provider is used
    • Self-hosted: Bare-metal
  • Kubernetes: (kubectl version)
    • Client: 1.11.1
    • Server: 1.11.1
  • Helm: (helm version)
    • Client: 2.11.0
    • Server: 2.11.0

Relevant logs

+ /scripts/set-config /var/opt/gitlab/templates /srv/gitlab/config
Begin parsing .erb files from /var/opt/gitlab/templates
Writing /srv/gitlab/config/gitlab.yml
Writing /srv/gitlab/config/resque.yml
Writing /srv/gitlab/config/database.yml
Copying other config files found in /var/opt/gitlab/templates
Copying smtp_settings.rb into /srv/gitlab/config
Copying unicorn.rb into /srv/gitlab/config
+ exec /scripts/wait-for-deps
Checking database connection and schema version
WARNING: This version of GitLab depends on gitlab-shell 8.4.3, but you're running Unknown. Please update gitlab-shell.
Database Schema
Current version: 0
Codebase version: 20181204154019
Checking database connection and schema version
Edited by silv