Skip to content

fix some issues in e2e tests #621

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.

Sign up for GitHub

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

Merged
merged 2 commits into from
Apr 6, 2025
Merged

fix some issues in e2e tests #621

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a8e77ee
added timeout to curl command which may otherwise hang
nirrozenbaum Mar 31, 2025
caa3282
check HF_TOKEN set at the beginning of the test
nirrozenbaum Mar 31, 2025
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
30 changes: 18 additions & 12 deletions test/e2e/epp/e2e_suite_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,8 @@ const (
defaultReadyTimeout = 3 * time.Minute
// defaultModelReadyTimeout is the default timeout for the model server deployment to report a ready state.
defaultModelReadyTimeout = 10 * time.Minute
// defaultCurlTimeout is the default timeout for the curl command to get a response.
defaultCurlTimeout = 30 * time.Second
// defaultInterval is the default interval to check if a resource exists or ready conditions.
defaultInterval = time.Millisecond * 250
// defaultCurlInterval is the default interval to run the test curl command.
Expand Down Expand Up @@ -107,7 +109,11 @@ var _ = ginkgo.BeforeSuite(func() {
})

func setupInfra() {
modelServerManifest := readModelServerManifestPath()
modelServerManifestPath := readModelServerManifestPath()
modelServerManifestArray := getYamlsFromModelServerManifest(modelServerManifestPath)
if strings.Contains(modelServerManifestArray[0], "hf-token") {
createHfSecret(cli, modelServerSecretManifest)
}
crds := map[string]string{
"inferencepools.inference.networking.x-k8s.io": inferPoolManifest,
"inferencemodels.inference.networking.x-k8s.io": inferModelManifest,
Expand All @@ -117,7 +123,7 @@ func setupInfra() {
createClient(cli, clientManifest)
createEnvoy(cli, envoyManifest)
// Run this step last, as it requires additional time for the model server to become ready.
createModelServer(cli, modelServerSecretManifest, modelServerManifest)
createModelServer(cli, modelServerManifestArray, modelServerManifestPath)
}

var _ = ginkgo.AfterSuite(func() {
Expand All @@ -137,7 +143,7 @@ func setupSuite() {
err = apiextv1.AddToScheme(scheme)
gomega.ExpectWithOffset(1, err).NotTo(gomega.HaveOccurred())

err = infextv1a2.AddToScheme(scheme)
err = infextv1a2.Install(scheme)
gomega.ExpectWithOffset(1, err).NotTo(gomega.HaveOccurred())

cli, err = client.New(cfg, client.Options{Scheme: scheme})
Expand Down Expand Up @@ -171,6 +177,7 @@ var (
existsTimeout = getTimeout("EXISTS_TIMEOUT", defaultExistsTimeout)
readyTimeout = getTimeout("READY_TIMEOUT", defaultReadyTimeout)
modelReadyTimeout = getTimeout("MODEL_READY_TIMEOUT", defaultModelReadyTimeout)
curlTimeout = getTimeout("CURL_TIMEOUT", defaultCurlTimeout)
interval = defaultInterval
curlInterval = defaultCurlInterval
)
Expand All @@ -191,6 +198,13 @@ func readModelServerManifestPath() string {
return modelServerManifestFilepath
}

func getYamlsFromModelServerManifest(modelServerManifestPath string) []string {
ginkgo.By("Ensuring the model server manifest points to an existing file")
modelServerManifestArray := readYaml(modelServerManifestPath)
gomega.Expect(modelServerManifestArray).NotTo(gomega.BeEmpty())
return modelServerManifestArray
}

// createCRDs creates the Inference Extension CRDs used for testing.
func createCRDs(k8sClient client.Client, crds map[string]string) {
for name, path := range crds {
Expand Down Expand Up @@ -224,15 +238,7 @@ func createClient(k8sClient client.Client, filePath string) {
}

// createModelServer creates the model server resources used for testing from the given filePaths.
func createModelServer(k8sClient client.Client, secretPath, deployPath string) {
ginkgo.By("Ensuring the model server manifest points to an existing file")
modelServerManifestArray := readYaml(deployPath)
gomega.Expect(modelServerManifestArray).NotTo(gomega.BeEmpty())
modelServerManifestYaml := modelServerManifestArray[0]
if strings.Contains(modelServerManifestYaml, "hf-token") {
createHfSecret(k8sClient, secretPath)
}

func createModelServer(k8sClient client.Client, modelServerManifestArray []string, deployPath string) {
ginkgo.By("Creating model server resources from manifest: " + deployPath)
createObjsFromYaml(k8sClient, modelServerManifestArray)

Expand Down
8 changes: 6 additions & 2 deletions test/e2e/epp/e2e_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,9 @@ package epp

import (
"fmt"
"strconv"
"strings"
"time"

"github.com/google/go-cmp/cmp"
"github.com/google/go-cmp/cmp/cmpopts"
Expand Down Expand Up @@ -53,7 +55,7 @@ var _ = ginkgo.Describe("InferencePool", func() {
}, existsTimeout, interval).Should(gomega.Succeed())

ginkgo.By("Verifying connectivity through the inference extension")
curlCmd := getCurlCommand(envoyName, nsName, envoyPort, modelName)
curlCmd := getCurlCommand(envoyName, nsName, envoyPort, modelName, curlTimeout)

// Ensure the expected responses include the inferencemodel target model names.
var expected []string
Expand Down Expand Up @@ -112,10 +114,12 @@ func newInferenceModel(ns string) *v1alpha2.InferenceModel {

// getCurlCommand returns the command, as a slice of strings, for curl'ing
// the test model server at the given name, namespace, port, and model name.
func getCurlCommand(name, ns, port, model string) []string {
func getCurlCommand(name, ns, port, model string, timeout time.Duration) []string {
return []string{
"curl",
"-i",
"--max-time",
strconv.Itoa((int)(timeout.Seconds())),
fmt.Sprintf("%s.%s.svc:%s/v1/completions", name, ns, port),
"-H",
"Content-Type: application/json",
Expand Down