E2E: verify RTE metrics endpoint TLS compliance#4259
Open
mrniranjan wants to merge 2 commits into
Open
Conversation
Add e2e tests that probe each RTE pod's metrics port directly over port-forward to confirm: - the TLS handshake succeeds and the negotiated version meets the cluster TLS profile minimum - the /metrics endpoint returns a valid response body - connections using a version or cipher incompatible with the cluster profile are rejected Add FetchMetrics helper in test/internal/tls to perform an HTTPS GET /metrics over a port-forwarded connection, reusing the existing port-forward and TLS probing infrastructure. AI Attribution: AIA Human-AI blend, New content, Human-initiated, Reviewed, Cursor 3.0.16 v1.0 Signed-off-by: Niranjan M.R <mniranja@redhat.com>
Contributor
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: mrniranjan The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-ci
Bot
added
the
approved
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Enterprise Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughAdds cluster-TLS-derived helpers and a single-connection HTTPS metrics fetcher, then updates E2E tests to validate RTE pod /metrics endpoints negotiate at-or-above the cluster minimum TLS and reject below-minimum versions or disallowed ciphers when applicable. ChangesRTE Metrics TLS Compliance Testing
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (1)
test/internal/tls/tls.go (1)
256-275: ⚡ Quick winAdd a request timeout to the metrics HTTP client.
The client and request (
context.Background()) have no deadline, sohttpCli.Do/io.ReadAllover the port-forward can block indefinitely if the server stalls. The table node has noSpecTimeout, so a hang would only be interrupted at suite level. A client-level timeout bounds this cheaply.♻️ Bound the request with a timeout
return &http.Client{Transport: tr} + return &http.Client{Transport: tr, Timeout: 30 * time.Second}Add the import:
"net/http" "strings" + "time"🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@test/internal/tls/tls.go` around lines 256 - 275, The metrics HTTP client returned by buildMetricsHTTPClient lacks any deadline and can hang; set a client-level timeout (e.g., assign the Timeout field on the returned *http.Client) so all requests via that client are bounded, and add the time import; update buildMetricsHTTPClient to construct &http.Client{Transport: tr, Timeout: <reasonable duration>} so http.Client and buildMetricsHTTPClient (and the transport tr) enforce a request timeout.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@test/internal/tls/tls.go`:
- Around line 256-275: The metrics HTTP client returned by
buildMetricsHTTPClient lacks any deadline and can hang; set a client-level
timeout (e.g., assign the Timeout field on the returned *http.Client) so all
requests via that client are bounded, and add the time import; update
buildMetricsHTTPClient to construct &http.Client{Transport: tr, Timeout:
<reasonable duration>} so http.Client and buildMetricsHTTPClient (and the
transport tr) enforce a request timeout.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Enterprise
Run ID: cf30fd60-4c84-4d28-8e74-4645014ce7fe
📒 Files selected for processing (2)
test/e2e/tls/tls_test.gotest/internal/tls/tls.go
shajmakh
reviewed
May 29, 2026
Comment on lines
+168
to
+173
| const rteMetricsPort = "2112" | ||
| minVersion, profileSpec := getClusterTLSProfile(ctx) | ||
| rtePods := getRTEPodsFromNRO(ctx) | ||
| for i := range rtePods { | ||
| rtePod := &rtePods[i] | ||
| By(fmt.Sprintf("Validating TLS metrics behavior on pod %s/%s", rtePod.Namespace, rtePod.Name)) |
Member
There was a problem hiding this comment.
The amount of shared code between positive and negative is small, while verifyRTETLSPositive and verifyRTETLSVersionRejected+verifyRTETLSCipherRejected are having most of the verification.
I think the goal of describeTable is lost here and having It() spec per test would serve readability better. WDYT?
Contributor
Author
There was a problem hiding this comment.
agree, addressed in latest commit
shajmakh
reviewed
May 29, 2026
Comment on lines
+188
to
+200
| func getClusterTLSProfile(ctx context.Context) (uint16, configv1.TLSProfileSpec) { | ||
| GinkgoHelper() | ||
| profileSpec, err := ctrltls.FetchAPIServerTLSProfile(ctx, e2eclient.Client) | ||
| Expect(err).ToNot(HaveOccurred(), "unable to fetch TLS profile from API Server") | ||
| tlsConfigFn, unSupportedCiphers := ctrltls.NewTLSConfigFromProfile(profileSpec) | ||
| if len(unSupportedCiphers) > 0 { | ||
| klog.InfoS("TLS profile has unsupported ciphers", "ciphers", unSupportedCiphers) | ||
| } | ||
| cfg := &tls.Config{} | ||
| tlsConfigFn(cfg) | ||
| klog.InfoS("cluster TLS minimum version", "versxion", tls.VersionName(cfg.MinVersion)) | ||
| return cfg.MinVersion, profileSpec | ||
| } |
Member
There was a problem hiding this comment.
haven't this been used before by the existing tests? if so, can we make it an test/internal/tls/tls.go function ?
Contributor
Author
There was a problem hiding this comment.
addressed in latest commit
shajmakh
reviewed
May 29, 2026
Comment on lines
+225
to
+271
| func verifyRTETLSPositive(rtePod *corev1.Pod, port string, minVersion uint16) { | ||
| GinkgoHelper() | ||
|
|
||
| By("Verifying TLS Handshake succeeds and negotiated version meets the cluster profile") | ||
| // fetch TLS Version and the cipher supported by TLS version | ||
| gotVersion, gotCipherId, err := intls.ProbeTLSSettings(e2eclient.K8sClient, rtePod, port) | ||
| Expect(err).ToNot(HaveOccurred(), "TLS Handshake failed on pod %q", rtePod.Name) | ||
| klog.InfoS("negotiated TLS settings", "version", tls.VersionName(gotVersion), "cipher", tls.CipherSuiteName(gotCipherId)) | ||
| Expect(gotVersion).To(BeNumerically(">=", minVersion), | ||
| "negotiated version %s below minimum %s", tls.VersionName(gotVersion), tls.VersionName(minVersion)) | ||
| By("Fetching /metrics over TLS via port-fortward") | ||
| body, err := intls.FetchMetrics(e2eclient.K8sClient, rtePod, port) | ||
| Expect(err).ToNot(HaveOccurred(), "failed to fetch metrics from pod %q", rtePod.Name) | ||
| Expect(body).ToNot(BeEmpty(), "metrics respose body should not be empty") | ||
| Expect(string(body)).To(ContainSubstring("rte_noderesourcetopology_writes_total")) | ||
| } | ||
|
|
||
| func verifyRTETLSVersionRejected(rtePod *corev1.Pod, port string, minVersion uint16) { | ||
| GinkgoHelper() | ||
|
|
||
| belowMinVersion, err := intls.TLSVersionBelow(minVersion) | ||
| Expect(err).ToNot(HaveOccurred(), "cannot determine version below %s", tls.VersionName(minVersion)) | ||
|
|
||
| By(fmt.Sprintf("Verifying TLS %s is rejected", tls.VersionName(belowMinVersion))) | ||
| err = intls.ProbeMaxTLSVersion(e2eclient.K8sClient, rtePod, port, belowMinVersion) | ||
| Expect(err).To(HaveOccurred(), "should reject TLS %s", tls.VersionName(belowMinVersion)) | ||
| Expect(errors.Is(err, intls.ErrTLSHandshakeRejected)).To(BeTrue(), "got: %v", err) | ||
| } | ||
|
|
||
| func verifyRTETLSCipherRejected(rtePod *corev1.Pod, port string, minVersion uint16, profileSpec configv1.TLSProfileSpec) { | ||
| GinkgoHelper() | ||
|
|
||
| if minVersion == tls.VersionTLS13 { | ||
| klog.InfoS("TLS 1.3 ciphers are not individually configurable, skipping cipher rejection test") | ||
| return | ||
| } | ||
|
|
||
| disallowedCipher := intls.FindDisallowedCipher(profileSpec.Ciphers) | ||
| if disallowedCipher == "" { | ||
| Skip("all known TLS 1.2 ciphers are in the allowed set, nothing to test") | ||
| } | ||
|
|
||
| By(fmt.Sprintf("Verifying cipher %s is rejected", disallowedCipher)) | ||
| err := intls.ProbeTLSCipher(e2eclient.K8sClient, rtePod, port, disallowedCipher) | ||
| Expect(err).To(HaveOccurred(), "should reject cipher %s", disallowedCipher) | ||
| Expect(errors.Is(err, intls.ErrTLSHandshakeRejected)).To(BeTrue(), "got: %v", err) | ||
| } |
Member
There was a problem hiding this comment.
if these are used only once let's have them inlined please.
Contributor
Author
There was a problem hiding this comment.
addressed in latest commit
Move getClusterTLSProfile to internal test utils function as GetClusterTLSProfile tls.Config logic moved to test/internal/tls/tls.go scheduler negative test By now names the disallowed cipher (e.g. ECDHE-ECDSA-AES128-SHA256) AI Attribution: AIA Human-AI blend, New content, Human-initiated, Reviewed, Cursor 3.0.16 v1.0 Signed-off-by: Niranjan M.R <mniranja@redhat.com>
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Add e2e tests that probe each RTE pod's metrics port directly over port-forward to confirm:
Add FetchMetrics helper in test/internal/tls to perform an HTTPS GET /metrics over a port-forwarded connection, reusing the existing port-forward and TLS probing infrastructure.
AI Attribution: AIA Human-AI blend, New content, Human-initiated, Reviewed, Cursor 3.0.16 v1.0