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

Split /tekton/run directories into separate volumes. #4278

Merged
merged 1 commit into from
Oct 11, 2021

Conversation

wlynch
Copy link
Member

@wlynch wlynch commented Oct 5, 2021

Changes

This changes splits each /tekton/run volume into separate volumes
so that steps can only mutate their own runtime information. This
prevents steps from unexpectedly interfering with other step execution.

To do this, we repurpose the /tekton/run/# path into a
step-specific directory. Since this was previously used by the
entrypoint for the post/wait files, we now use /tekton/run/#/out as
the post/wait filepath instead.

This does not change behavior of the directory, it enforces expected
behavior of steps.

/tekton/run is considered an internal implementation detail and is not
covered by the API compatibility policy, so it is safe to make changes
to the behavior of these files/paths.

This does not stop user execution from writing the step's own /tekton/run/#
folder. This needs more discussion/design - additional changes (if
needed) will be made in another commit.

This change is only focused on /tekton/run to reduce PR complexity. We will
likely want to make a similar change to /tekton/steps in another commit.
We may also look to consolidate all per-step volumes into a single
source (i.e. creds-init does something similar as well).

AFAICT, Ephemeral Volumes (i.e. EmptyDir) are exempt from Node Volume
limits (https://kubernetes.io/docs/concepts/storage/storage-limits/) -
spot checked this with a TaskRun with 100+ steps on both kind and GKE.

Part of #4227

/kind cleanup

Performance

Not very rigorous/scientific, but I ran 50 TaskRuns against this commit and v0.28.1 on a kind cluster, each with 100+ steps of:

    - image: ubuntu
      script: echo hello
Version v0.29.1 This PR
Avg 0:01:50 0:01:59
Median 0:01:50 0:01:58
Min 0:01:46 0:01:53
Max 0:02:04 0:02:28

Submitter Checklist

As the author of this PR, please check off the items in this checklist:

  • Docs included if any changes are user facing
  • Tests included if any functionality added or changed
  • Follows the commit message standard
  • Meets the Tekton contributor standards (including
    functionality, content, code)
  • Release notes block below has been filled in or deleted (only if no user facing changes)

Release Notes

NONE

Internal only changes.

@tekton-robot tekton-robot added release-note Denotes a PR that will be considered when it comes time to generate release notes. kind/cleanup Categorizes issue or PR as related to cleaning up code, process, or technical debt. labels Oct 5, 2021
@tekton-robot tekton-robot added the size/L Denotes a PR that changes 100-499 lines, ignoring generated files. label Oct 5, 2021
@tekton-robot
Copy link
Collaborator

The following is the coverage report on the affected files.
Say /test pull-tekton-pipeline-go-coverage to re-run this coverage report

File Old Coverage New Coverage Delta
pkg/pod/entrypoint.go 87.0% 86.8% -0.2
pkg/pod/pod.go 90.7% 91.2% 0.5

@tekton-robot tekton-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Oct 7, 2021
pkg/pod/entrypoint_test.go Outdated Show resolved Hide resolved
@ghost
Copy link

ghost commented Oct 7, 2021

Would it be worthwhile to add either an e2e test or an example yaml that exercises the RO behaviour of the volumes?

E.g. something like:

kind: TaskRun
apiVersion: tekton.dev/v1beta1
spec:
  taskSpec:
    steps:
    - image: # image with a shell
      script: |
        set +e # dont fail the script on error
        printf "hello world" > /tekton/steps/run/1/foo.txt
        if [ $? -eq 0 ] ; then
          echo "able to write to run directory of non-current step"
          exit 1
        fi
      - image: # this step exists only to have the run/1 volumemount in the first step
        script: echo "step 2"

@tekton-robot tekton-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Oct 8, 2021
@wlynch
Copy link
Member Author

wlynch commented Oct 8, 2021

Would it be worthwhile to add either an e2e test or an example yaml that exercises the RO behaviour of the volumes?

E.g. something like:

kind: TaskRun
apiVersion: tekton.dev/v1beta1
spec:
  taskSpec:
    steps:
    - image: # image with a shell
      script: |
        set +e # dont fail the script on error
        printf "hello world" > /tekton/steps/run/1/foo.txt
        if [ $? -eq 0 ] ; then
          echo "able to write to run directory of non-current step"
          exit 1
        fi
      - image: # this step exists only to have the run/1 volumemount in the first step
        script: echo "step 2"

Done. (I wish there was a separate test-only directory, but added to examples for now)

@tekton-robot
Copy link
Collaborator

The following is the coverage report on the affected files.
Say /test pull-tekton-pipeline-go-coverage to re-run this coverage report

File Old Coverage New Coverage Delta
pkg/pod/entrypoint.go 87.0% 86.8% -0.2
pkg/pod/pod.go 87.8% 88.5% 0.7

pkg/pod/pod_test.go Outdated Show resolved Hide resolved
This changes splits each `/tekton/run` volume into separate volumes
so that steps can only mutate their own runtime information. This
prevents steps from unexpectedly interfering with other step execution.

To do this, we repurpose the `/tekton/run/#` path into a
step-specific directory. Since this was previously used by the
entrypoint for the post/wait files, we now use `/tekton/run/#/out` as
the post/wait filepath instead.

This does not change behavior of the directory, it enforces expected
behavior of steps.

`/tekton/run` is considered an internal implementation detail and is not
covered by the API compatibility policy, so it is safe to make changes
to the behavior of these files/paths.

This does not stop user execution from writing the step's own `/tekton/run/#`
folder. This needs more discussion/design - additional changes (if
needed) will be made in another commit.

This change is only focused on `/tekton/run` to reduce PR complexity. We will
likely want to make a similar change to /tekton/steps in another commit.
We may also look to consolidate all per-step volumes into a single
source (i.e. creds-init does something similar as well).

AFAICT, Ephemeral Volumes (i.e. EmptyDir) are exempt from Node Volume
limits (https://kubernetes.io/docs/concepts/storage/storage-limits/) -
spot checked this with a TaskRun with 100+ steps on both kind and GKE.
@tekton-robot
Copy link
Collaborator

The following is the coverage report on the affected files.
Say /test pull-tekton-pipeline-go-coverage to re-run this coverage report

File Old Coverage New Coverage Delta
pkg/pod/entrypoint.go 87.0% 86.8% -0.2
pkg/pod/pod.go 87.8% 88.5% 0.7

@ghost
Copy link

ghost commented Oct 8, 2021

Cheers!

/lgtm

@tekton-robot tekton-robot assigned ghost Oct 8, 2021
@tekton-robot tekton-robot added the lgtm Indicates that a PR is ready to be merged. label Oct 8, 2021
@wlynch
Copy link
Member Author

wlynch commented Oct 8, 2021

/test pull-tekton-pipeline-alpha-integration-tests pull-tekton-pipeline-integration-tests

@wlynch
Copy link
Member Author

wlynch commented Oct 8, 2021

/test pull-tekton-pipeline-integration-tests

Copy link
Member

@vdemeester vdemeester left a comment

Choose a reason for hiding this comment

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

/retest

@tekton-robot
Copy link
Collaborator

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: vdemeester

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

The pull request process is described 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

@tekton-robot tekton-robot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Oct 11, 2021
@tekton-robot tekton-robot merged commit 00f7cc6 into tektoncd:main Oct 11, 2021
@lbernick lbernick mentioned this pull request Jun 17, 2022
3 tasks
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. kind/cleanup Categorizes issue or PR as related to cleaning up code, process, or technical debt. lgtm 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. size/L Denotes a PR that changes 100-499 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants