Long timeout how to debug

Hi,
We have 0.8.4 of drone server running in the same lab as Mr. robertlabrie ( my cubemate ) and I kicked off a drone build to just run a docker build. here is the exact .drone.yaml file ( I was building a non master branch)

#
# Build_and_publish_master will only build and push
# when the github repository has a new tag set
#
# Build_not_master: should build when not the master branch
#
#
pipeline:
  build_and_publish_master:
    image: plugins/docker
    repo: tropo/kubernetes-github-authn
    username: ${dockerhub_user}
    password: ${dockerhub_pass}
    default_tags: true
    when:
      event: [tag]
    branch: master

  build_not_master:
    image: docker
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    commands:
      - docker build . 

And there was a significant timeout that happened for some reason that I don’t understand. looking at the build log:


there is a ~ 4 1/4 minute where no activity was taking place. The docker logs on the agent box doesn’t show anything and on the server log there is not even anmention of the build. I would also say that running drone exec the build runs just fine.

I’ve also experienced builds being “stuck” during the run.

So far I have no proper solution for debugging, but for us it seems that this correlates with IO issues - when there is too much IOPs processes just stall for some time.

Any other hints on debugging occurrences like this are highly appreciated