Do not log potentially sensitive data below DEBUG log level

[pierDipi]

Currently EPP and BBR might log potentially sensitive data at DEFAULT or VERBOSE log level.

Part of #1162

[netlify]

✅ Deploy Preview for gateway-api-inference-extension ready!

Name	Link
🔨 Latest commit	76d1859
🔍 Latest deploy log	https://app.netlify.com/projects/gateway-api-inference-extension/deploys/6880df6acf298e0008c6fd07
😎 Deploy Preview	https://deploy-preview-1192--gateway-api-inference-extension.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[k8s-ci-robot]

Hi @pierDipi. Thanks for your PR.

I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[kfswain]

/ok-to-test

Looks great, thanks! Will stamp when tests pass.

[pierDipi]

Removed to avoid double logs for the same problem

{"level":"error","ts":"2025-07-23T11:07:49Z","caller":"handlers/server.go:210","msg":"Error unmarshaling request body","x-request-id":"3c11cbb0-b7ed-4699-b823-069026d18386","error":"invalid character 'D' after object key:value pair","stacktrace":"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/handlers.(*StreamingServer).Process\n\t/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/epp/handlers/server.go:210\ngithub.com/envoyproxy/go-control-plane/envoy/service/ext_proc/v3._ExternalProcessor_Process_Handler\n\t/go/pkg/mod/github.com/envoyproxy/go-control-plane/[email protected]/service/ext_proc/v3/external_processor_grpc.pb.go:106\ngoogle.golang.org/grpc.(*Server).processStreamingRPC\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1695\ngoogle.golang.org/grpc.(*Server).handleStream\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1819\ngoogle.golang.org/grpc.(*Server).serveStreams.func2.1\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1035"}
{"level":"error","ts":"2025-07-23T11:07:49Z","caller":"handlers/server.go:310","msg":"Failed to process request","x-request-id":"3c11cbb0-b7ed-4699-b823-069026d18386","request":"request_body:{end_of_stream:true}  metadata_context:{}","error":"inference gateway: BadRequest - Error unmarshaling request body: {\n  \"model\": \"facebook/opt-125m\",\n  \"prompt\": \"\"Delve into the multifaceted implications of a fully disaggregated cloud architecture, specifically where the compute plane (P) and the data plane (D) are independently deployed and managed for a geographically distributed, high-throughput, low-latency microservices ecosystem. Beyond the fundamental challenges of network latency and data consistency, elaborate on the advanced considerations and trade-offs inherent in such a setup: 1. Network Architecture and Protocols: How would the network fabric and underlying protocols (e.g., RDMA, custom transport layers) need to evolve to support optimal performance and minimize inter-plane communication overhead, especially for synchronous operations? Discuss the role of network programmability (e.g., SDN, P4) in dynamically optimizing routing and traffic flow between P and D. 2. Advanced Data Consistency and Durability: Explore sophisticated data consistency models (e.g., causal consistency, strong eventual consistency) and their applicability in balancing performance and data integrity across a globally distributed data plane. Detail strategies for ensuring data durability and fault tolerance, including multi-region replication, intelligent partitioning, and recovery mechanisms in the event of partial or full plane failures. 3. Dynamic Resource Orchestration and Cost Optimization: Analyze how an orchestration layer would intelligently manage the independent scaling of compute (P) and data (D) resources, considering fluctuating workloads, cost efficiency, and performance targets (e.g., using predictive analytics for resource provisioning). Discuss mechanisms for dynamically reallocating compute nodes to different data partitions based on workload patterns and data locality, potentially involving live migration strategies. 4. Security and Compliance in a Distributed Landscape: Address the enhanced security perimeter challenges, including securing communication channels between P and D (encryption in transit, mutual TLS), fine-grained access control to data at rest and in motion, and identity management across disaggregated components. Discuss how such an architecture impacts compliance with regulatory frameworks (e.g., GDPR, HIPAA) concerning data sovereignty, privacy, and auditability. 5. Operational Complexity and Observability: Examine the increased complexity in monitoring, logging, and tracing across highly decoupled compute and data planes. What specialized tooling and practices (e.g., distributed tracing with OpenTelemetry, advanced AIOps) would be essential? How would incident response and troubleshooting differ in this disaggregated environment compared to traditional integrated systems? Consider the challenges of pinpointing root causes across independent failures. 6. Real-world Applicability and Future Trends: Identify specific industries or use cases (e.g., high-frequency trading, IoT edge processing, large language model inference) where the benefits of P/D disaggregation would strongly outweigh its complexities. Conclude by speculating on emerging technologies or paradigms (e.g., serverless compute functions directly interacting with object storage, in-memory disaggregation) that could further drive or transform P/D disaggregation in cloud computing.\", \n  \"stream\": false, \n  \"max_tokens\": 50}","stacktrace":"sigs.k8s.io/gateway-api-inference-extension/pkg/epp/handlers.(*StreamingServer).Process\n\t/go/pkg/mod/sigs.k8s.io/[email protected]/pkg/epp/handlers/server.go:310\ngithub.com/envoyproxy/go-control-plane/envoy/service/ext_proc/v3._ExternalProcessor_Process_Handler\n\t/go/pkg/mod/github.com/envoyproxy/go-control-plane/[email protected]/service/ext_proc/v3/external_processor_grpc.pb.go:106\ngoogle.golang.org/grpc.(*Server).processStreamingRPC\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1695\ngoogle.golang.org/grpc.(*Server).handleStream\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1819\ngoogle.golang.org/grpc.(*Server).serveStreams.func2.1\n\t/go/pkg/mod/google.golang.org/[email protected]/server.go:1035"}

[kfswain]

sgtm

[kfswain]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: kfswain, pierDipi

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [kfswain]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment