Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

os/Dockerfile: Use dnf, pin repos, and clean cache #23613

Open
wants to merge 1 commit into
base: master
from

Conversation

@ashcrow
Copy link
Member

commented Aug 15, 2019

  • Clean the dnf/yum cache before running commands
  • Pin Fedora repos
  • Call dnf directly instead of yum
@openshift-ci-robot

This comment has been minimized.

Copy link

commented Aug 15, 2019

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: ashcrow
To complete the pull request process, please assign smarterclayton
You can assign the PR to them by writing /assign @smarterclayton in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@openshift-ci-robot openshift-ci-robot requested review from pweil- and sdodson Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

Error getting repository data for built, repository not found
@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

I'm looking to see where built.repo is defined. I'm wondering if it may have been accidentally removed somewhere but is still expected by this build.

@ashcrow ashcrow force-pushed the ashcrow:clean-cache-and-use-f30 branch from 59cc9aa to 0896ba7 Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

Looks like f29 required in the name for other dependencies in the CI set up. Switched back to f29 to see if it passes.

@ashcrow ashcrow changed the title WIP: os/Dockerfile: Use F30, dnf, and clean cache WIP: os/Dockerfile: Use dnf and clean cache Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

Looking to pin repos instead of using mirrors.

os/Dockerfile: Use dnf, pin yum repos, and clean cache
- Clean the dnf/yum cache before running commands
- Call dnf directly instead of yum
- Pin to a specific set of Fedora repos

@ashcrow ashcrow force-pushed the ashcrow:clean-cache-and-use-f30 branch from 0896ba7 to 7fb6654 Aug 15, 2019

@openshift-ci-robot openshift-ci-robot added size/M and removed size/XS labels Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

/test e2e-cmd

@ashcrow ashcrow changed the title WIP: os/Dockerfile: Use dnf and clean cache os/Dockerfile: Use dnf and clean cache Aug 15, 2019

@ashcrow ashcrow changed the title os/Dockerfile: Use dnf and clean cache os/Dockerfile: Use dnf, pin repos, and clean cache Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

/test e2e-aws

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

The image builds fine, just need to get past the other CI targets.

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 15, 2019

/cc @jlebon

@openshift-ci-robot openshift-ci-robot requested a review from jlebon Aug 15, 2019

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 16, 2019

/test e2e-cmd

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 16, 2019

no such image 😕

/test e2e-cmd

@jlebon

This comment has been minimized.

Copy link
Member

commented Aug 16, 2019

Is the motivation here using mirrors because they can flake? If a build observes many of them failing in a single run, there might something else wrong which using a specific mirror won't fix.

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 16, 2019

@jlebon more or less yes. The Fedora mirrors have been flaky at random times causing significant failures.

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 16, 2019

s: "error waiting for the pod 'test-cmd' to complete: the pod errored trying to run the command",

/test e2e-cmd

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 19, 2019

/test e2e-cmd

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 20, 2019

Failed:

[Suite:openshift/test-cmd][Serial][Disruptive] test-cmd: test/cmd/policy-storage-admin.sh [Suite:openshift] expand_less

/test e2e-cmd

@cgwalters

This comment has been minimized.

Copy link
Contributor

commented Aug 20, 2019

Ultimately though, if we're going to use Fedora content in CI, we need to make the mirroring more robust, not work around it at every call site.

That said another way to work around this would be changing the system to use coreos-assembler instead like openshift/machine-config-operator#682

@ashcrow

This comment has been minimized.

Copy link
Member Author

commented Aug 20, 2019

Ultimately though, if we're going to use Fedora content in CI, we need to make the mirroring more robust, not work around it at every call site.

True. This PR is attempting to work around the pain for the short term, not a forever fix.

@vrutkovs

This comment has been minimized.

Copy link
Contributor

commented Sep 10, 2019

FAILURE after 0.000s: test/cmd/basicresources.sh:175: executing 'oc set probe -f /var/tests/test/cmd/testdata/hello-openshift/hello-pod.json --local -o yaml --liveness --get-url=https://127.0.0.1:port/path' expecting success and text 'port: port': the command returned the wrong error code; the output content test failed
There was no output from the command.
Standard error from the command:
error: --get-url could not be parsed as a valid URL: parse https://127.0.0.1:port/path: invalid port ":port" after host
[ERROR] hack/lib/cmd.sh:241: `return "${return_code}"` exited with status 1.

/retest

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
5 participants
You can’t perform that action at this time.