integration tests running with updated aws-k8s-tester #766
Conversation
| if [[ "$PROVISION" = true ]]; then | ||
| up-test-cluster | ||
| __cluster_created=1 |
jaypipes
force-pushed
the
e2e-test-cleanup
branch
2 times, most recently
from
12bdad1
to
f784b4a
Compare
jaypipes
changed the title
Adds some guardrails to the scripts/run-integration_tests.sh script by checking (before creating an EKS cluster) that AWS credentials are present and that the required ECR repository is readable.
Ensure that there is an ECR repo before we use aws-k8s-tester to actually create the test cluster. In addition, trap errors in the main run-integration-tests.sh script and deprovision cluster if an error occurred after creating the cluster. Updates the required aws-k8s-tester release to 0.5.2 to handle aws/aws-k8s-tester#66 Issue aws#765
jaypipes
force-pushed
the
e2e-test-cleanup
branch
from
f784b4a
to
5b7a46f
Compare
jaypipes
changed the title
| # login to the ECR registry | ||
| eval $(aws ecr get-login --region $AWS_REGION --no-include-email) >/dev/null 2>&1 | ||
| ensure_ecr_repo "$AWS_ACCOUNT_ID" "$AWS_ECR_REPO_NAME" | ||
| make docker IMAGE=$IMAGE_NAME VERSION=$IMAGE_VERSION |
There was a problem hiding this comment.
This doesn't pull source code for a given tag(version) right?
There was a problem hiding this comment.
No, it doesn't. It runs docker build -t aws-vpc-cni:$IMAGE_VERSION. If IMAGE_VERSION envvar has been overridden, we should probably set the BUILD envvar to false automatically.
There was a problem hiding this comment.
Either that, or check that IMAGE_VERSION is overridden and if so, do a git checkout $IMAGE_VERSION here to switch the working tree to that git tag.
There was a problem hiding this comment.
using git-checkout makes sense. Can you help me understand how disabling BUILD will help? Do you mean to say, if CNI image is already available in my repo then disabling BUILD will help?
There was a problem hiding this comment.
yes, exactly. if you've already pushed, say, a v1.5.5 CNI image built from that git tag/branch, then setting BUILD=false will prevent the run-integration-tests.sh script from docker pushing an image and the eks-k8s-tester binary will simply pull the image that was in the ECR repo for v1.5.5
There was a problem hiding this comment.
Ah ok makes sense. Thanks for clarifying.
There was a problem hiding this comment.
In short, I need to make some changes to this script to facilitate the not-building-from-local-checkout scenarios...
jaypipes
force-pushed
the
e2e-test-cleanup
branch
from
5b7a46f
to
3fb53de
Compare
The names of environment variables accepted by aws-k8s-tester changed when the managed node group functionality was introduced. This commit updates the integration testing scripts to call aws-k8s-tester (v.0.5.4 which is the release needed with the fix for aws/aws-k8s-tester#70) with these updated environment variables. We decrease the number of parallel builds of the echo job from 100 to 3 and the number of completions for that job from 1000 to 30. This decreases the setup time of the cluster by about 10 minutes. Finally, I added in a short-circuit to prevent double-deprovisioning of the cluster if, say, a stacktrace occurred when running the aws-k8s-tester tool. Issue aws#686 Issue aws#784 Issue aws#786
jaypipes
force-pushed
the
e2e-test-cleanup
branch
from
3fb53de
to
cf66674
Compare
Checks that there is an ECR repo before we use aws-k8s-tester to actually
create the test cluster. In addition, trap errors in the main
run-integration-tests.sh script and deprovision cluster if an error
occurred after creating the cluster.
Adds some guardrails to the scripts/run-integration_tests.sh script by
checking (before creating an EKS cluster) that AWS credentials are
present and that the required ECR repository is readable.
The names of environment variables accepted by aws-k8s-tester changed
when the managed node group functionality was introduced. This commit
updates the integration testing scripts to call aws-k8s-tester (v.0.5.4
which is the release needed with the fix for aws/aws-k8s-tester#70) with
these updated environment variables.
We decrease the number of parallel builds of the echo job from 100 to
3 and the number of completions for that job from 1000 to 30. This
decreases the setup time of the cluster by about 10 minutes.
Finally, I added in a short-circuit to prevent double-deprovisioning of
the cluster if, say, a stacktrace occurred when running the
aws-k8s-tester tool.
Issue #686
Issue #784
Issue #786
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.