e2e-test: enable real TEE on self-hosted runners #284
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
ok we can't merge this yet, since the e2e test is scheduled on the PR's fork (
see this: #223 (comment) let's do those changes to the workflow in the scope of this PR. sample e2e can still run anywhere |
mkulke
force-pushed
the
mkulke/enable-az-snp-e2e-test
branch
from
8208cbd
to
887379a
Compare
mkulke
changed the title
Xynnn007
reviewed
Jan 12, 2024
| @@ -0,0 +1,43 @@ | |||
| #!/usr/bin/env bash | |||
| # | |||
| # Copyright (c) 2023 Intel Corporation | |||
There was a problem hiding this comment.
it's copied from CAA, which copied it from kata afaik, so I left the header in
We want to be able to test PRs in a real TEE on self-hosted runners. By default the tests are executed in the fork from which the PR is crated. The fork does not have the self-hosted runners which provide the TEE. We need to run them in the scope of the target repository in a safe fashion that prevents untrusted PRs from running arbitrary code on self-hosted runners. This change is modelled after simular workflows in kata and CAA repos. The matrix-based kbs-e2e workflow has been turned into a reusable workflow that is being called by specific TEE workflows. In this PR that would be the "sample" and "az-snp-vtpm" TEE. For a real TEE we follow the following ceremony: 1) verify whether the PR is authorized to run (a maintainer needs to set a label on the PR). 2) checkout the main branch from the target 3) rebase the changes from the source atop the main branch (using a script copied verbatim from CAA) 4) archive the resulting HEAD and upload it as artifact 5) call the re-usable workflow with the required runners param 6) extract the artifact 7) run the test For the sample TEE steps 1-3 are replaced by a simple checkout Signed-off-by: Magnus Kulke <[email protected]>
mkulke
force-pushed
the
mkulke/enable-az-snp-e2e-test
branch
from
887379a
to
54606ce
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Why
We want to be able to test PRs in a real TEE on self-hosted runners. By
default the tests are executed in the fork from which the PR is crated.
The fork does not have the self-hosted runners which provide the TEE. We
need to run them in the scope of the target repository in a safe fashion
that prevents untrusted PRs from running arbitrary code on self-hosted
runners.
fixes: #223
How
This change is modelled after simular workflows in kata and CAA repos.
The matrix-based kbs-e2e workflow has been turned into a reusable
workflow that is being called by specific TEE workflows. In this PR that
would be the "sample" and "az-snp-vtpm" TEE.
For a real TEE we follow the following ceremony:
a label on the PR).
script copied verbatim from CAA)
For the sample TEE steps 1-3 are replaced by a simple checkout
Caveats
This yaml spaghetti is very hard to test. So while this passes actionlint,
it might need several iterations to always work.
The reusable workflow is targetting main and we cannot point it to a
fork, so no workflow run will successfully complete in this PR.