Sorry Cypress Director Pods dies

[Lory1990]

Hi! I am running sorry Cypress with Gitlab Pipielines, here is the pipeline:

download-libraries:
  image: cypress/browsers:node18.6.0-chrome105-ff104
  timeout: 180m
  allow_failure: false
  cache:
    paths:
      - node_modules/
      - .npm
      - .cache/Cypress
      - cache/Cypress
  script:
    - npm install
    - npx cypress cache path
    - npx cypress cache list
    - npx cypress verify
    - npx cypress info
  artifacts:
    paths:
      - .npm
      - .cache/Cypress
      - cache/Cypress
      - node_modules

test-on-sorry-cypress:
  image: cypress/browsers:node18.6.0-chrome105-ff104
  timeout: 180m
  when: on_success
  parallel: 5
  allow_failure: true
  needs:
    - download-libraries
  tags:
    - cypress-runner
  variables:
    FF_USE_FASTZIP: "true"
  script:
    - npx cy2 run --record --key ${CYPRESS_RECORD_KEY} --browser chrome --ci-build-id ${CI_PIPELINE_IID} --group ${CYPRESS_RECORD_KEY} --parallel
  artifacts:
    when: always
    paths:
      - cypress/videos/**/*.mp4
      - cypress/screenshots/**/*.png
    expire_in: 1 day

The runners with tag cypress-runner are pods deployed in Azure Kubernetes Services and seems that some of them "dies" before starting the run with this message:

$ npx cy2 run --record --key ${CYPRESS_RECORD_KEY} --browser chrome --ci-build-id ${CI_PIPELINE_IID} --group ${CYPRESS_RECORD_KEY} --parallel
[cy2] Running cypress with API URL: https://cypress-api.step4business.com/
[54:1103/050854.505865:ERROR:zygote_host_impl_linux.cc(263)] Failed to adjust OOM score of renderer with pid 230: Permission denied (13)
libva error: vaGetDriverNameByIndex() failed with unknown libva error, driver_name = (null)
[230:1103/050855.911187:ERROR:gpu_memory_buffer_support_x11.cc(44)] dri3 extension not supported.
Couldn't find tsconfig.json. tsconfig-paths will be skipped

====================================================================================================
  (Run Starting)
[.....]
====================================================================================================

ERROR: Job failed: pod "runner-rxduhhn-project-28738-concurrent-0x9b94" status is "Failed"

and this is totally random... do you have the same issue?

[agoldis]

That seems like cypress runner's failure - are there any more details?

[Lory1990]

Unfortunately not, the only bad log I can see is:

{"level":50,"time":1667562393323,"requestId":"c9da59d5-e881-4c6d-9574-09882b51fc0a","error":{"type":"Error","message":"connect ECONNREFUSED 127.0.0.1:9100","stack":"Error: connect ECONNREFUSED 127.0.0.1:9100\n    at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1247:16)\n    at TCPConnectWrap.callbackTrampoline (node:internal/async_hooks:130:17)","errno":-111,"code":"ECONNREFUSED","syscall":"connect","address":"127.0.0.1","port":9100},"instanceId":"5ccb7343-8780-4937-82f3-6421bb53e788","msg":"Unable to get upload instructions"}

Could this help?

[Lory1990]

After a day of tesing i made MINIO work and i got this error

Timed out waiting for the browser to connect. Retrying...
[118](integration-tests/-/jobs/1137494#L118)Missing browserCriClient in connectToNewSpec
[119](integration-tests/-/jobs/1137494#L119)Error: Missing browserCriClient in connectToNewSpec
[120](integration-tests/-/jobs/1137494#L120)    at Object.connectToNewSpec (/builds/clients/beconsulting/step/step/services/step-integration-tests/cache/Cypress/10.11.0/Cypress/resources/app/packages/server/lib/browsers/chrome.js:451:19)
[121](integration-tests/-/jobs/1137494#L121)    at Object.connectToNewSpec (/builds/clients/beconsulting/step/step/services/step-integration-tests/cache/Cypress/10.11.0/Cypress/resources/app/packages/server/lib/browsers/index.js:96:31)
[122](integration-tests/-/jobs/1137494#L122)    at processTicksAndRejections (node:internal/process/task_queues:96:5)
[123](integration-tests/-/jobs/1137494#L123)    at OpenProject.relaunchBrowser (/builds/clients/beconsulting/step/step/services/step-integration-tests/cache/Cypress/10.11.0/Cypress/resources/app/packages/server/lib/open_project.js:150:24)

[Lory1990]

Hi everyone sorry for the delay, I solved the issue!

The problem was that Kubernetes could not give to the pod enough memory so the pod freezes and dies.

Solved adding more memory to Kubernetes (size up machine pool)