e2e-testing

name

e2e-testing

description

Run Camunda e2e/smoke tests against a deployed cluster — generate .env credentials with render-e2e-env.sh or deploy-camunda, run Playwright suites via c8e2e (distributed on-cluster runner) or --test-e2e, test multiple environments in parallel, and reproduce CI e2e failures locally. Use when running or debugging e2e/smoke tests, generating test credentials for a namespace, or reproducing a failing CI e2e job.

Running E2E Tests

Throughout: $NS = Kubernetes namespace, $RELEASE = Helm release name.

Generate a per-environment .env file

After deploying a namespace, generate the .env.test file that the Playwright test suite needs. This file contains the ingress hostname, resolved credentials (from Kubernetes secrets), Keycloak URLs, and feature flags — all auto-resolved from the live cluster.

Via deploy-camunda:

# Generate .env.test alongside the deployment
deploy-camunda \
  --chart-path ./charts/camunda-platform-8.9 \
  --namespace $NS --release $RELEASE \
  --scenario chart-full-setup --identity keycloak --persistence elasticsearch \
  --output-test-env

# Custom path (useful for multiple environments side by side)
deploy-camunda ... --output-test-env --output-test-env-path .env.test.eske

For multi-scenario parallel deployments, each scenario gets its own .env.test.{scenario} file automatically.

Standalone against an existing namespace:

./scripts/render-e2e-env.sh \
  --absolute-chart-path $PWD/charts/camunda-platform-8.9 \
  --namespace $NS \
  --output .env.test \
  --kube-context $CTX        # optional: target a specific cluster
  # --opensearch             # set IS_OPENSEARCH=true
  # --rba                    # set IS_RBA=true
  # --mt                     # set IS_MT=true
  # --run-smoke-tests        # set IS_SMOKE=true
  # -v                       # verbose: show resolved values

Run the tests

Via deploy-camunda (deploy + test in one step):

deploy-camunda \
  --chart-path ./charts/camunda-platform-8.9 \
  --namespace $NS --release $RELEASE \
  --scenario chart-full-setup --identity keycloak --persistence elasticsearch \
  --test-e2e          # e2e tests after deploy
  # --test-it         # integration tests instead
  # --test-all        # both

Via c8e2e (distributed Playwright runner on Kubernetes):

c8e2e (@camunda/c8e2e) launches sharded Playwright test pods directly on the cluster — faster and more reliable than running locally. Point it at a deployed environment by its ingress URL:

# Run against a deployed namespace
c8e2e test \
  --target SM-8.9 \
  --endpoint https://$NS.ci.distro.ultrawombat.com \
  --feature-flags smoke \
  --follow

# Full test suite with sharding
c8e2e test \
  --target SM-8.9 \
  --endpoint https://my-env.ci.distro.ultrawombat.com \
  --shards 4 \
  --follow

# Filter to specific tests
c8e2e test \
  --target SM-8.9 \
  --endpoint https://my-env.ci.distro.ultrawombat.com \
  --grep "Basic Navigation"

# OpenSearch environment with multitenancy
c8e2e test \
  --target SM-8.9 \
  --endpoint https://os-env.ci.distro.ultrawombat.com \
  --feature-flags opensearch,mt \
  --follow

Manage running tests:

c8e2e list                    # List active test runs
c8e2e status <run-id>         # Check status
c8e2e logs <run-id>           # Stream logs
c8e2e results <run-id>        # Download results
c8e2e cancel <run-id>         # Cancel a run

Multiple environments in parallel

Deploy multiple namespaces with unique subdomains, then run c8e2e against each:

# Deploy two environments with unique hostnames
deploy-camunda --chart-path ./charts/camunda-platform-8.9 \
  --namespace test-es --release camunda --ingress-subdomain test-es \
  --identity keycloak --persistence elasticsearch

deploy-camunda --chart-path ./charts/camunda-platform-8.9 \
  --namespace test-os --release camunda --ingress-subdomain test-os \
  --identity keycloak --persistence opensearch-embedded

# Run e2e against each in parallel
c8e2e test --target SM-8.9 --endpoint https://test-es.ci.distro.ultrawombat.com --feature-flags smoke --follow &
c8e2e test --target SM-8.9 --endpoint https://test-os.ci.distro.ultrawombat.com --feature-flags smoke,opensearch --follow &
wait

Reproducing a CI Test Failure Locally

See docs/skills/reproducing-ci-e2e-failures.md for the step-by-step guide to pulling logs, downloading artifacts, decoding CI scenario shortnames, and spinning up an identical local environment.

For the full deploy → credentials → reproduce-on-main → verify-on-branch loop, see the gke-verification skill.