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

Revert to using isolated PID namespaces in Docker #51634

Merged
merged 2 commits into from
Sep 6, 2017

Conversation

verb
Copy link
Contributor

@verb verb commented Aug 30, 2017

What this PR does / why we need it: Reverts to the previous docker default of using isolated PID namespaces for containers in a pod. There exist container images that expect always to be PID 1 which we want to support unmodified in 1.8.

Which issue this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close that issue when PR gets merged): fixes #48937

Special notes for your reviewer:

Release note:

Sharing a PID namespace between containers in a pod is disabled by default in 1.8. To enable for a node, use the --docker-disable-shared-pid=false kubelet flag. Note that PID namespace sharing requires docker >= 1.13.1.

@k8s-ci-robot k8s-ci-robot added size/XS Denotes a PR that changes 0-9 lines, ignoring generated files. cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. labels Aug 30, 2017
@k8s-github-robot k8s-github-robot added the release-note Denotes a PR that will be considered when it comes time to generate release notes. label Aug 30, 2017
@verb
Copy link
Contributor Author

verb commented Aug 30, 2017

looks like maybe the e2e node test doesn't set the flag. Will look into it tomorrow.

@derekwaynecarr
Copy link
Member

poke when tests pass, thanks for making the update.

@dims
Copy link
Member

dims commented Aug 31, 2017

@verb do we want to specify the docker versions needed for the shared pid flag to work in the release note?

@verb
Copy link
Contributor Author

verb commented Aug 31, 2017

@dims oh yes, great catch. Updated the release note.

@verb
Copy link
Contributor Author

verb commented Aug 31, 2017

@yguo0905 can you offer any advice about how to best to update the node e2e test for this? It looks like the kubelet is only run one time for the test, so I can't test with both the flag enabled and disabled.

Should I test the default behavior or test the feature even though it's disabled by default?

I assume the former but I want to make sure I'm not missing something.

@yguo0905
Copy link
Contributor

I think we want to test this behavior because user can enable it if they want. We can set the flag at

KUBELET_ARGS='--cgroups-per-qos=true --cgroup-root=/'

@k8s-ci-robot k8s-ci-robot added size/S Denotes a PR that changes 10-29 lines, ignoring generated files. and removed size/XS Denotes a PR that changes 0-9 lines, ignoring generated files. labels Sep 1, 2017
@verb
Copy link
Contributor Author

verb commented Sep 1, 2017

/retest

1 similar comment
@verb
Copy link
Contributor Author

verb commented Sep 1, 2017

/retest

@@ -155,6 +155,7 @@ func (e *E2EServices) startKubelet() (*server, error) {
"--serialize-image-pulls", "false",
"--pod-manifest-path", manifestPath,
"--file-check-frequency", "10s", // Check file frequently so tests won't wait too long
"--docker-disable-shared-pid=false",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we set it here we don't need the change in test/e2e_node/jenkins/jenkins-pull.properties?

Actually, if we don't set it here there are other properties files in test/e2e_node/jenkins/ need to be changed.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, thanks. Based on this advice (and because it made the e2e node test pass) I've removed the jenkins config and set the flag here.

verb added 2 commits September 1, 2017 23:50
A shared PID namespace were enabled by default in the 1.7 when running
with a supported Docker runtime, but a Docker version that supports
a shared namespace was not qualified for use.

Release 1.8 will qualify a docker version supporting shared PID, but we
don't want to cause disruption for container images which expect always
to have PID 1.
This also renames isSharedPIDNamespaceEnabled() to
isSharedPIDNamespaceSupported() to be more accurate.
@verb verb force-pushed the sharedpid-default-off branch from 48f203e to 765374c Compare September 1, 2017 21:51
@verb
Copy link
Contributor Author

verb commented Sep 2, 2017 via email

@verb
Copy link
Contributor Author

verb commented Sep 2, 2017

/retest

1 similar comment
@verb
Copy link
Contributor Author

verb commented Sep 4, 2017

/retest

@verb
Copy link
Contributor Author

verb commented Sep 5, 2017

This PR is ready for review. According to sig-testing, pull-kubernetes-e2e-kops-aws is broken and has been disabled for the merge queue.

This PR should be added to the 1.8 milestone to resolve critical-urgent issue #48937.

@dims
Copy link
Member

dims commented Sep 5, 2017

@kubernetes/sig-release-members Can you please apply milestone v1.8?

@ericchiang ericchiang added this to the v1.8 milestone Sep 5, 2017
@derekwaynecarr
Copy link
Member

/lgtm
/approve

@k8s-ci-robot k8s-ci-robot added the lgtm "Looks good to me", indicates that a PR is ready to be merged. label Sep 5, 2017
@k8s-github-robot
Copy link

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: derekwaynecarr, verb

Associated issue: 48937

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

Needs approval from an approver in each of these OWNERS Files:

You can indicate your approval by writing /approve in a comment
You can cancel your approval by writing /approve cancel in a comment

@k8s-github-robot k8s-github-robot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Sep 5, 2017
@verb
Copy link
Contributor Author

verb commented Sep 5, 2017

/retest

@fejta-bot
Copy link

/retest
This bot automatically retries jobs that failed/flaked on approved PRs (send feedback to @fejta).

Review the full test history for this PR.

@k8s-ci-robot
Copy link
Contributor

@verb: The following test failed, say /retest to rerun them all:

Test name Commit Details Rerun command
pull-kubernetes-e2e-kops-aws 765374c link /test pull-kubernetes-e2e-kops-aws

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

@k8s-github-robot
Copy link

Automatic merge from submit-queue (batch tested with PRs 51984, 51351, 51873, 51795, 51634)

@k8s-github-robot k8s-github-robot merged commit eb86cc5 into kubernetes:master Sep 6, 2017
@jdumars
Copy link
Member

jdumars commented Sep 6, 2017

@verb could you please assign a SIG to this?

@verb
Copy link
Contributor Author

verb commented Sep 7, 2017

/sig node

@k8s-ci-robot k8s-ci-robot added the sig/node Categorizes an issue or PR as relevant to SIG Node. label Sep 7, 2017
chaitanyaenr added a commit to chaitanyaenr/svt that referenced this pull request Nov 28, 2017
In 1.7 release the shared namespace is enabled by default. This is
causing a problem when running pbench container which runs systemd.

Related issue in upstream kubernetes:
kubernetes/kubernetes#48937

According to my understanding this will be fixed from 1.8 with the
following patch:
kubernetes/kubernetes#51634
chaitanyaenr added a commit to chaitanyaenr/svt that referenced this pull request Nov 28, 2017
In 1.7 release the shared PID namespaces is enabled by default. This is
causing a problem when running pbench container which runs systemd.

Related issue in upstream kubernetes:
kubernetes/kubernetes#48937

According to my understanding this will be fixed from 1.8 with the
following patch:
kubernetes/kubernetes#51634
mffiedler pushed a commit to openshift/svt that referenced this pull request Nov 29, 2017
In 1.7 release the shared PID namespaces is enabled by default. This is
causing a problem when running pbench container which runs systemd.

Related issue in upstream kubernetes:
kubernetes/kubernetes#48937

According to my understanding this will be fixed from 1.8 with the
following patch:
kubernetes/kubernetes#51634
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
approved Indicates a PR has been approved by an approver from all required OWNERS files. cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. lgtm "Looks good to me", indicates that a PR is ready to be merged. release-note Denotes a PR that will be considered when it comes time to generate release notes. sig/node Categorizes an issue or PR as relevant to SIG Node. size/S Denotes a PR that changes 10-29 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Cannot deploy GlusterFS through Kubernetes - "Couldn't find an alternative telinit implementation to spawn"
10 participants