Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions hack/test-e2e.sh
Original file line number Diff line number Diff line change
Expand Up @@ -50,5 +50,6 @@ else
fi
fi

kubectl delete crd inferencepools.inference.networking.k8s.io --ignore-not-found
Copy link
Contributor

Choose a reason for hiding this comment

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

why this script deletes the CRD?

Copy link
Author

Choose a reason for hiding this comment

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

This may be a timing issue in my local environment. Aborting or failing tests (moreso the former) followed by immediate re-execution sees an InferencePool that hasn't been torn down yet. I would imagine other devs experience this as well so I added it, but I can remove it if not a good practice.

echo "Found an active cluster. Running Go e2e tests in ./epp..."
go test ./test/e2e/epp/ -v -ginkgo.v
6 changes: 4 additions & 2 deletions test/e2e/epp/e2e_suite_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,7 @@ import (

const (
// defaultCurlTimeout is the default timeout for the curl command to get a response.
defaultCurlTimeout = 30 * time.Second
defaultCurlTimeout = 120 * time.Second
Copy link
Contributor

Choose a reason for hiding this comment

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

2 min for curl timeout is a bit long.
why do we need to increase the 30 sec timeout?

Copy link
Author

Choose a reason for hiding this comment

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

I think you're right, though 30 was too short. When generating traffic in verifyMetrics I was getting timeouts (curl 28). I implemented retries but was still getting timeouts with 30s -- only increasing the timeout worked. I am not confident that increasing the number of retries alone would resolve it. I am open to suggestions.

// defaultCurlInterval is the default interval to run the test curl command.
defaultCurlInterval = time.Second * 5
// defaultNsName is the default name of the Namespace used for tests. Can override using the E2E_NS environment variable.
Expand Down Expand Up @@ -317,7 +317,9 @@ func createEnvoy(testConfig *testutils.TestConfig, filePath string) {

// createInferExt creates the inference extension resources used for testing from the given filePath.
func createInferExt(testConfig *testutils.TestConfig, filePath string) {
inManifests := testutils.ReadYaml(filePath)

// This image needs to be updated to open multiple ports and respond.
inManifests := testutils.ReadYaml(filePath) // Modify inference-pool.yaml
ginkgo.By("Replacing placeholders with environment variables")
outManifests := []string{}
replacer := strings.NewReplacer(
Expand Down
Loading