# Remote Procedure Call (RPC)

Remote procedure call (RPC) is a client-server communication model in which a caller sends procedure parameters to a server and receives procedure results in a reply.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## How to aggregate test results across teams with Temporal Nexus

DevFeed: [How to aggregate test results across teams with Temporal Nexus](<https://devfeed.tech/articles/how-to-aggregate-test-results-across-teams-with-temporal-nexus-146865.md>)

Original publisher: [Read original article](<https://temporal.io/blog/how-to-aggregate-test-results-across-teams-with-temporal-nexus>)

Author: Brooks Newberry

Published: 2026-11-20T00:00:00Z

Content type: article

Language: en

Sources: [Temporal](<https://devfeed.tech/sources/temporal-blog.md>)

Topics: [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Integration testing](<https://devfeed.tech/topics/integration-testing.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>)

Tags: [aggregate](<https://devfeed.tech/tags/aggregate.md>), [collect](<https://devfeed.tech/tags/collect.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [format](<https://devfeed.tech/tags/format.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [nexus](<https://devfeed.tech/tags/nexus.md>), [release](<https://devfeed.tech/tags/release.md>), [release-engineering](<https://devfeed.tech/tags/release-engineering.md>), [report](<https://devfeed.tech/tags/report.md>), [results](<https://devfeed.tech/tags/results.md>), [teams](<https://devfeed.tech/tags/teams.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [temporal-voices](<https://devfeed.tech/tags/temporal-voices.md>), [test](<https://devfeed.tech/tags/test.md>)

### AI overview

Temporal's Release Engineering team built a shared report format and Nexus Operation to collect test evidence from cloud validation and open source test Workflows while leaving tests and verdicts with their owners. Producers call a named Operation through an allowed Temporal Endpoint using their existing Temporal connections; a direct gRPC write API would require a reachable service and separate authentication and access controls. The synchronous handler validates and stores reports, keyed by (source, business_id), then returns the record ID. That key makes retries after a lost response return the same record, but repeated submissions can replace contents, and older payloads are not rejected by timestamp. Each report and result uses pass, warn, fail, or skipped; a test failure can coexist with successful ingestion. The team reads reports through a separate gRPC API. Nexus access is limited to explicitly allowed Namespaces in the same Cloud Account, and the synchronous database work must finish within ten seconds; Temporal recommends asynchronous Operations for uncertain database latency or availability. Reports support release decisions but do not automatically determine readiness,

### Source excerpt

How Temporal's Release Engineering team used a Nexus Operation and a shared report format to collect test results from many teams in one place.

## CVE-2026-129448 - Apache Roller: Unauthenticated deserialization in the XML-RPC endpoint (releases before 6.1.5)

DevFeed: [CVE-2026-129448 - Apache Roller: Unauthenticated deserialization in the XML-RPC endpoint (releases before 6.1.5)](<https://devfeed.tech/articles/cve-2026-129448-apache-roller-unauthenticated-deserialization-in-the-xml-rpc-endpoint-releases-before-6-1-5-136957.md>)

Original publisher: [Read original article](<https://cvefeed.io/vuln/detail/CVE-2026-129448>)

Author: cvefeed.io

Published: 2026-11-17T11:03:49Z

Content type: news

Language: en

Sources: [Latest High/Critical Vulnerabilitiy Feed](<https://devfeed.tech/sources/latest-high-critical-vulnerabilitiy-feed.md>)

Topics: [cve remediation](<https://devfeed.tech/topics/cve-remediation.md>), [Data serialization](<https://devfeed.tech/topics/data-serialization.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [XML](<https://devfeed.tech/topics/xml.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [actively-exploited](<https://devfeed.tech/tags/actively-exploited.md>), [apache](<https://devfeed.tech/tags/apache.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cve-2026-129448](<https://devfeed.tech/tags/cve-2026-129448.md>), [deserialization](<https://devfeed.tech/tags/deserialization.md>), [exploited](<https://devfeed.tech/tags/exploited.md>), [roller](<https://devfeed.tech/tags/roller.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [xml](<https://devfeed.tech/tags/xml.md>)

### AI overview

Apache Roller 4.0.0 through 6.1.4 permits unauthenticated remote code execution through its XML-RPC endpoint: it deserializes vendor extension types before authenticating requests. The issue is reported as actively exploited. Upgrade to 6.1.6 or later, which disables those extension types and rejects XML-RPC requests when the feature is disabled. If an upgrade is not possible, set `enabledForExtensions` to `false` in `WEB-INF/web.xml` or remove the `/roller-services/xmlrpc` servlet mapping, then restart Roller. Disabling XML-RPC in the administration UI alone is insufficient on older versions because deserialization occurs before that setting is checked. The source also says this is the same vulnerability as CVE-2026-82384, previously listed as affecting only 6.1.5.

### Source excerpt

CVE ID :CVE-2026-129448 Published : Nov. 17, 2026, 11:03 a.m. | 20 minutes ago Description :Deserialization of Untrusted Data in Apache Roller versions 4.0.0 through 6.1.4 allows an unauthenticated remote attacker to execute arbitrary code. The XML-RPC endpoint accepts vendor extension types and deserializes them before authenticating the requester. This is the same vulnerability as CVE-2026-82384, which listed only Apache Roller 6.1.5 as affected. This issue is being actively exploited. Users are recommended to upgrade to Apache Roller 6.1.6 or later, which disables the extension types and rejects XML-RPC requests when the XML-RPC feature is disabled. Users who cannot upgrade right away should set "enabledForExtensions" to "false" in WEB-INF/web.xml, or remove the "/roller-services/xmlrpc" servlet mapping, and restart Roller. Disabling XML-RPC in the administration UI is not enough, because older versions deserialize the request before checking that setting. Severity: 9.8 | CRITICAL Visit the link for more details, such as CVSS details, affected products, timeline, and more...

## Clean-Looking Addresses Still Fail Delivery. Here Is How to Catch Them in Snowflake

DevFeed: [Clean-Looking Addresses Still Fail Delivery. Here Is How to Catch Them in Snowflake](<https://devfeed.tech/articles/clean-looking-addresses-still-fail-delivery-here-is-how-to-catch-them-in-snowflake-129820.md>)

Original publisher: [Read original article](<https://medium.com/snowflake/clean-looking-addresses-still-fail-delivery-here-is-how-to-catch-them-in-snowflake-32cd51d5f4c4?source=rss----34b6daafc07---4>)

Author: Jan Rehorik

Published: 2026-11-11T20:01:01Z

Content type: tutorial

Language: en

Sources: [Snowflake Builders](<https://devfeed.tech/sources/snowflake-builders.md>)

Topics: [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>)

Tags: [address-validation](<https://devfeed.tech/tags/address-validation.md>), [api](<https://devfeed.tech/tags/api.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [batch](<https://devfeed.tech/tags/batch.md>), [boundary](<https://devfeed.tech/tags/boundary.md>), [carrier](<https://devfeed.tech/tags/carrier.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [dedupe](<https://devfeed.tech/tags/dedupe.md>), [encrypted](<https://devfeed.tech/tags/encrypted.md>), [external](<https://devfeed.tech/tags/external.md>), [fields](<https://devfeed.tech/tags/fields.md>), [formatting](<https://devfeed.tech/tags/formatting.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [install](<https://devfeed.tech/tags/install.md>), [integration](<https://devfeed.tech/tags/integration.md>), [join](<https://devfeed.tech/tags/join.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [mailing](<https://devfeed.tech/tags/mailing.md>), [marketplace](<https://devfeed.tech/tags/marketplace.md>), [route](<https://devfeed.tech/tags/route.md>), [snowflake](<https://devfeed.tech/tags/snowflake.md>)

### AI overview

The Snowflake app validates US addresses against postal reference data through a stored procedure that reads a supplied SELECT, sends selected address fields to the Service Objects API, and writes results to a named table. The excerpt's workflow maps seven required input columns, runs a batch with one CALL, then uses DPV codes and errors to route records--for example, accept DPV 1, flag DPV 3 or 4 for unit review, and investigate other outcomes. Original source values can remain intact for review and rollback. Data leaves the account through an approved external access integration, is sent over HTTPS, and the app says it retains nothing. The 90-day or 1,000-record trial requires external access; Snowflake trial accounts may need that capability enabled. Validation confirms deliverability, not resident identity or whether a customer has moved.

### Source excerpt

Clean-Looking Addresses Still Fail Delivery. Here Is How to Catch ThemHow to validate and standardize US addresses in Snowflake using SQL One source writes "26 S Chestnut St." Another writes "26 s chstnt, vntura, californa" with the wrong ZIP code. Both are the same building in Ventura, and both will break a join, a dedupe rule, or a mailing run. Formatting rules in SQL can make strings look alike. They cannot tell you whether an address is a real delivery point, or whether it is missing the suite number the carrier needs. That takes validation against postal reference data, and in most stacks it means an export, an API client, and a re-import to maintain. In this tutorial, you'll use SQL to submit addresses for validation, review the results, and decide what happens to each record. You will: Install Service Objects' US Address Validation and Verification for Snowflake app from the Marketplace. Shape your address columns into the input the app expects, without changing your source table. Validate a batch of records with one CALL. Read the results and route each record with SQL. Every screenshot and result below comes from a real run in a Snowsight worksheet. Architecture DiagramThe app runs in your account. Only the address fields you select cross the boundary, through one integration you approve. The app is installed into your Snowflake account and exposes a stored procedure. You pass it a SELECT statement and an output table. It reads those rows, sends the address fields to the Service Objects API for validation, and writes the results into the table you named. From there, the results are ordinary Snowflake data. The external access integration is the one point where data leaves your account. The app sends address data to the Service Objects API through a single external access integration you grant and revoke, and retains nothing. Data is encrypted in transit over HTTPS. That boundary is worth showing your security reviewer up front. Why formatting rules are not

## fRAC: Remote Accelerator Calls on Hostless FPGAs

DevFeed: [fRAC: Remote Accelerator Calls on Hostless FPGAs](<https://devfeed.tech/articles/frac-remote-accelerator-calls-on-hostless-fpgas-127829.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2504.04404>)

Author: Ziyi Yang, Krishnan B. Iyer, Yixi Chen, Ran Shu, Zsolt Istv\\'an, Marco Canini, Suhaib A. Fahmy

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [rdma](<https://devfeed.tech/topics/rdma.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [dpu](<https://devfeed.tech/tags/dpu.md>), [fpgas](<https://devfeed.tech/tags/fpgas.md>), [functions](<https://devfeed.tech/tags/functions.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [interface](<https://devfeed.tech/tags/interface.md>), [latency](<https://devfeed.tech/tags/latency.md>), [optimized](<https://devfeed.tech/tags/optimized.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [power](<https://devfeed.tech/tags/power.md>), [prototype](<https://devfeed.tech/tags/prototype.md>), [remote](<https://devfeed.tech/tags/remote.md>), [request](<https://devfeed.tech/tags/request.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [throughput](<https://devfeed.tech/tags/throughput.md>), [transport](<https://devfeed.tech/tags/transport.md>), [under](<https://devfeed.tech/tags/under.md>)

### AI overview

fRAC moves network-to-accelerator invocation into FPGA hardware: it handles transport ingestion, request reassembly, dispatch, and accelerator launch without host software. The prototype is a lightweight shim on an FPGA hosting multiple virtualized accelerators, which can be reconfigured in a few milliseconds. The authors report under 7.8 μs end-to-end latency and up to 90 Gbps application throughput. Against optimized DPDK-based CPU and DPU baselines, fRAC reports at least 80% lower latency under saturation and 85% less power. The abstract identifies serverless functions using fine-grained accelerators as a suitable use case because host invocation overhead can match or exceed their execution time.

### Source excerpt

arXiv:2504.04404v3 Announce Type: replace Abstract: Hardware accelerators are essential for meeting microsecond-scale demands in distributed applications, but invoking them over the network remains bottlenecked by host software. Host-mediated paths (RPC, RDMA) still rely on the CPU to reassemble requests and launch accelerators, which dominates end-to-end latency, while packet-level offload on SmartNICs lacks any notion of a complete request. We introduce the Remote Accelerator Call (RAC) paradigm, in which transport ingestion, request reassembly, dispatch, and accelerator invocation are handled entirely in hardware. We build fRAC, a hostless prototype that implements this network-to-accelerator interface as a lightweight shim on an FPGA hosting multiple virtualized accelerators, reconfigurable at runtime in a few milliseconds. Serverless functions backed by fine-grained accelerators are a natural fit, since host-mediated invocation overhead can match or exceed their short execution times. fRAC achieves under 7.8us end-to-end latency while sustaining application throughput of up to 90 Gbps. Compared to highly optimized DPDK-based CPU and DPU baselines, it achieves at least 80% lower latency under saturation, while consuming 85% less power.

## Retrying a send is harder than sending

DevFeed: [Retrying a send is harder than sending](<https://devfeed.tech/articles/retrying-a-send-is-harder-than-sending-113381.md>)

Original publisher: [Read original article](<https://www.sitepoint.com/retrying-a-send-is-harder-than-sending/>)

Author: Sadaf Ali

Published: 2026-11-03T04:47:03Z

Content type: opinion

Language: en

Sources: [SitePoint](<https://devfeed.tech/sources/sitepoint.md>)

Topics: [client](<https://devfeed.tech/topics/client.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Concurrency](<https://devfeed.tech/topics/concurrency.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [boolean](<https://devfeed.tech/tags/boolean.md>), [database](<https://devfeed.tech/tags/database.md>), [deduplication](<https://devfeed.tech/tags/deduplication.md>), [delivered](<https://devfeed.tech/tags/delivered.md>), [delivery](<https://devfeed.tech/tags/delivery.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [false-positives](<https://devfeed.tech/tags/false-positives.md>), [idempotency](<https://devfeed.tech/tags/idempotency.md>), [integration](<https://devfeed.tech/tags/integration.md>), [job](<https://devfeed.tech/tags/job.md>), [messaging](<https://devfeed.tech/tags/messaging.md>), [model](<https://devfeed.tech/tags/model.md>), [mutating](<https://devfeed.tech/tags/mutating.md>), [technology](<https://devfeed.tech/tags/technology.md>)

### AI overview

A timed-out send leaves the caller unable to tell whether the provider acted, so retrying can duplicate an external side effect. The article says reliable deduplication requires a caller-generated idempotency key reused on every attempt, with the provider storing and returning the original result for a defined window; local tracking and searching recent jobs cannot reliably solve a distributed ambiguity. It also argues that idempotency does not reveal delivery status: Fax for Agents uses queued, active, partial, failed, unknown and delivered states. Unknown calls for waiting, partial should not be automatically resent, and delivered is reserved for provider evidence beyond mere acceptance. The article recommends checking API documentation for key scope and duration, job states and guidance for unknown outcomes, and webhook signature and event-ID verification.

### Source excerpt

null Continue reading Retrying a send is harder than sending on SitePoint.

## Building a Simple MCP Server for KubeDesk (Open-Source Kubernetes GUI)

DevFeed: [Building a Simple MCP Server for KubeDesk (Open-Source Kubernetes GUI)](<https://devfeed.tech/articles/building-a-simple-mcp-server-for-kubedesk-open-source-kubernetes-gui-109720.md>)

Original publisher: [Read original article](<https://blog.devops.dev/a-simple-mcp-server-implementation-example-with-open-source-kubernetes-management-gui-kubedesk-b73a80e2144e?source=rss----33f8b2d9a328---4>)

Author: Yasar Gozudeli

Published: 2026-10-30T18:59:53Z

Content type: tutorial

Language: en

Sources: [DevOps.dev](<https://devfeed.tech/sources/devops-dev.md>)

Topics: [MSP MCP](<https://devfeed.tech/topics/msp-mcp.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [rancher](<https://devfeed.tech/topics/rancher.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [adapter](<https://devfeed.tech/tags/adapter.md>), [agent](<https://devfeed.tech/tags/agent.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [desktop-app](<https://devfeed.tech/tags/desktop-app.md>), [devops](<https://devfeed.tech/tags/devops.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [generative-ai-tools](<https://devfeed.tech/tags/generative-ai-tools.md>), [gui](<https://devfeed.tech/tags/gui.md>), [implementation](<https://devfeed.tech/tags/implementation.md>), [java](<https://devfeed.tech/tags/java.md>), [kubectl](<https://devfeed.tech/tags/kubectl.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-protocol](<https://devfeed.tech/tags/mcp-protocol.md>), [mcp-server](<https://devfeed.tech/tags/mcp-server.md>), [model-context-protocol](<https://devfeed.tech/tags/model-context-protocol.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [pods](<https://devfeed.tech/tags/pods.md>), [server](<https://devfeed.tech/tags/server.md>), [simple](<https://devfeed.tech/tags/simple.md>)

### AI overview

The guide shows how to add a small, read-only MCP server to KubeDesk as an adapter over its existing ClusterService, letting coding agents inspect Kubernetes through the app's service layer. It uses the JDK HTTP server and Jackson already on the classpath, exposes tools for health scans and resource inspection, and documents a Streamable HTTP example using the 2025-11-25 protocol flow; the text notes that later revisions require their own transport rules. It recommends server-side filtering, compact and capped results, stable context and namespace defaults, argument validation, timeouts, bounded resource use, and actionable tool errors. Loopback binding alone does not secure local read access: Origin checks, an authentication decision, Kubernetes RBAC, and careful handling of sensitive logs and manifests matter. The first version omits write tools and remote access; a CLI can offer similar controls, while MCP provides a standard interface for compatible clients.

### Source excerpt

A practical guide to exposing an existing Java application through the Model Context Protocol, with no new framework and a deliberately small set of read-only tools. KubeDesk is a lightweight, open source desktop GUI for Kubernetes, a kind of kubectl without the terminal. In this article I am going to explain how an experimental, read-only MCP server implemented in the project.KubeDesk MCP Server panel How a devops engineer normally work with kubectl, or any other kubernetes management tool? they check some information in the cluster, sometimes in pods, sometimes in some other kubernetes item and then they copy some information to LLM Chat screen, and then LLM either ask them some additional information or gives some code to execute and the loop goes on. How it would be if an agent could directly connect to kubernetes management tool and use it under a human survalience? So in this article I am going to tell you how it can be implemented so that developers can have a basic idea of what is MCP from end to end and see it is in action. 1. The misconception to clear first When I described the idea, the first question might be: "Why do I even need a server? Can't the agent just call KubeDesk's code directly?" "KubeDesk is already a tool working on my desktop -- isn't this tools-of-a-tool(my local coding agent can call the source code) already?" The useful distinction is between the model deciding what to do and the application executing it. For this implementation, there is one MCP connection: Host = the application running the agent loop, such as Codex or Claude Code. It talks to the model and manages tool execution; the model itself may run remotely. Client = the MCP component inside that application running the agent loop. It discovers tools and sends tool calls. Server = the program exposing those tools and executing them. In KubeDesk, it is an adapter with no LLM of its own. That is our design choice, not a restriction imposed by MCP. Coding agents such as Codex and

## Task-Oriented Semantic Communication for Hazard Warning and Remote Operation in Connected Vehicle Platoons

DevFeed: [Task-Oriented Semantic Communication for Hazard Warning and Remote Operation in Connected Vehicle Platoons](<https://devfeed.tech/articles/task-oriented-semantic-communication-for-hazard-warning-and-remote-operation-in-connected-vehicle-platoons-107605.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.06242>)

Author: S M Sabit Bananee, Shahriar Hasan, Muhammad Mahbub Alam, Nafiul Rashid

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [format](<https://devfeed.tech/topics/format.md>)

Tags: [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [automated](<https://devfeed.tech/tags/automated.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [broadcast](<https://devfeed.tech/tags/broadcast.md>), [channel](<https://devfeed.tech/tags/channel.md>), [communication](<https://devfeed.tech/tags/communication.md>), [compression](<https://devfeed.tech/tags/compression.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [data](<https://devfeed.tech/tags/data.md>), [data-transmission](<https://devfeed.tech/tags/data-transmission.md>), [decision-making](<https://devfeed.tech/tags/decision-making.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [effect](<https://devfeed.tech/tags/effect.md>), [emergency](<https://devfeed.tech/tags/emergency.md>), [esm](<https://devfeed.tech/tags/esm.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [exchange](<https://devfeed.tech/tags/exchange.md>), [framework](<https://devfeed.tech/tags/framework.md>), [remote](<https://devfeed.tech/tags/remote.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [warning](<https://devfeed.tech/tags/warning.md>)

### AI overview

The proposed platoon communication framework combines onboard instance segmentation and monocular depth estimation for hazard decisions with SD-VAE semantic scene compression. It jointly broadcasts task-relevant hazard information and a compact scene representation in an Emergency Semantic Message over the existing V2X stack, aiming to avoid transmitting high-volume raw sensor data. On task-relevant reconstruction metrics, SD-VAE outperformed JPEG and JPEG 2000 and degraded more gradually at low SINR, supporting its feasibility for remote scene reconstruction when human intervention is needed. Across 7,290 simulations, the reported inter-vehicle collision rate was 0.99%; collisions occurred mainly at extreme speeds and short gaps.

### Source excerpt

arXiv:2609.06242v2 Announce Type: replace Abstract: Connected and automated vehicle platoons require reliable dissemination of task-relevant hazard information to enable appropriate downstream responses, with unresolved situations additionally requiring remote human intervention. However, the limited channel bandwidth of V2X communication makes high-volume sensor data transmission challenging, motivating the exchange of only task-relevant information rather than raw sensor data. This paper proposes an integrated semantic and task-oriented communication framework for cooperative hazard response and remote operation in automated vehicle platoons. The framework combines onboard instance segmentation and monocular depth estimation for task-oriented hazard decision-making with Stable Diffusion Variational Autoencoder (SD-VAE)-based semantic scene compression. Task-relevant hazard information and the compact semantic representation are jointly disseminated through a broadcast-based Emergency Semantic Message (ESM) over the existing V2X protocol stack. Evaluation demonstrates that SD-VAE consistently outperforms JPEG and JPEG 2000 on task-relevant reconstruction metrics, exhibiting graceful degradation rather than a sharp cliff effect under low-SINR conditions, demonstrating its feasibility for remote scene reconstruction when human intervention is required. Moreover, an extensive campaign of 7,290 simulation runs shows a low overall inter-vehicle collision rate of 0.99%, with collisions occurring mainly under extreme high-speed and short-gap conditions.

## We've built the fastest way to create a one-click remote MCP server by mistake

DevFeed: [We've built the fastest way to create a one-click remote MCP server by mistake](<https://devfeed.tech/articles/we-ve-built-the-fastest-way-to-create-a-one-click-remote-mcp-server-by-mistake-109522.md>)

Original publisher: [Read original article](<https://appwrite.io/blog/post/one-click-remote-mcp-server-by-mistake>)

Author: Eldad Fux

Published: 2026-10-30T00:00:00Z

Content type: article

Language: en

Sources: [Appwrite Blog](<https://devfeed.tech/sources/appwrite-blog.md>)

Topics: [Back end](<https://devfeed.tech/topics/backend.md>), [implementation](<https://devfeed.tech/topics/implementation.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [client](<https://devfeed.tech/topics/client.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [OpenID connect (OIDC)](<https://devfeed.tech/topics/oidc.md>), [Server-side rendering](<https://devfeed.tech/topics/server-side-rendering.md>), [OpenAPI Specification](<https://devfeed.tech/topics/openapi.md>)

Tags: [access-tokens](<https://devfeed.tech/tags/access-tokens.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-tool](<https://devfeed.tech/tags/ai-tool.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [api](<https://devfeed.tech/tags/api.md>), [auth](<https://devfeed.tech/tags/auth.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [email](<https://devfeed.tech/tags/email.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-server](<https://devfeed.tech/tags/mcp-server.md>), [oauth](<https://devfeed.tech/tags/oauth.md>), [production](<https://devfeed.tech/tags/production.md>), [remote-mcp-server](<https://devfeed.tech/tags/remote-mcp-server.md>)

### AI overview

An Appwrite project can host a remote MCP server with OAuth 2.1 authorization, consent, and user-scoped data access already integrated. The example server uses protected resource metadata and a 401 challenge for discovery, then verifies signed access tokens offline against the project's JWKS, checking issuer, the server-specific audience, token type, and scopes. Its tools are exposed only when the user's approved scopes permit them, with scope checks repeated on each call. The server can run as a Function on its own domain or as an SSR Site route; the example uses a Function and returns plain JSON per request, so streaming tools need a different design. The approach is fast because hosting, identity, and data share a project, though that consolidates those services on one platform. The OAuth2 server is currently an Appwrite Cloud feature, not available in self-hosted Appwrite. Revoking consent prevents new tokens, while already issued offline-verified tokens remain usable until expiry, one hour by default for public clients.

### Source excerpt

We built an email client on Appwrite to test a new feature, then asked an agent for a remote MCP server with OAuth 2.1. It was in production in under a minute.

## Snowflake's Managed MCP Server: Giving AI Assistants a Governed Door Into Your Warehouse (Part 1)

DevFeed: [Snowflake's Managed MCP Server: Giving AI Assistants a Governed Door Into Your Warehouse (Part 1)](<https://devfeed.tech/articles/snowflake-s-managed-mcp-server-giving-ai-assistants-a-governed-door-into-your-warehouse-part-1-102727.md>)

Original publisher: [Read original article](<https://medium.com/snowflake/snowflakes-managed-mcp-server-giving-ai-assistants-a-governed-door-into-your-warehouse-part-1-09077cd30c61?source=rss----34b6daafc07---4>)

Author: Eylon Steiner

Published: 2026-10-27T19:01:02Z

Content type: article

Language: en

Sources: [Snowflake Builders](<https://devfeed.tech/sources/snowflake-builders.md>)

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Query (disambiguation)](<https://devfeed.tech/topics/query.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-assistants](<https://devfeed.tech/tags/ai-assistants.md>), [analyst](<https://devfeed.tech/tags/analyst.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [chat](<https://devfeed.tech/tags/chat.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [claude](<https://devfeed.tech/tags/claude.md>), [client](<https://devfeed.tech/tags/client.md>), [console](<https://devfeed.tech/tags/console.md>), [container](<https://devfeed.tech/tags/container.md>), [context](<https://devfeed.tech/tags/context.md>), [cortex](<https://devfeed.tech/tags/cortex.md>), [managed](<https://devfeed.tech/tags/managed.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-server](<https://devfeed.tech/tags/mcp-server.md>), [server](<https://devfeed.tech/tags/server.md>), [snowflake](<https://devfeed.tech/tags/snowflake.md>), [snowflake-data-superhero](<https://devfeed.tech/tags/snowflake-data-superhero.md>), [warehouse](<https://devfeed.tech/tags/warehouse.md>)

### AI overview

Snowflake's managed MCP server exposes warehouse tools as schema-level objects, with calls running under users' Snowflake identities and existing RBAC controls. The example separates a Cortex Agent backed by a semantic view and ticket search, plus a freshness procedure, from a second server with read-only SQL for engineers; placing SQL beside the agent could let a client bypass the semantic layer. The test found that generic procedures must return one cell (the author changed a table result to VARIANT), and unrestricted PATs combined with secondary roles could expose tools beyond the intended role. The author recommends role-restricted PATs and OAuth with secondary roles set to NONE. Query history records calls with the user, role and MCP request ID. Limits include tools only, 50 tools per server, 250 KB truncation for generic and SQL results, and manual OAuth client registration. The OAuth setup was not tested in the sandbox.

### Source excerpt

One door in, behind your own RBAC: that's the whole idea of the Snowflake-managed MCP server.This is Part 1 of a two-part series. Here I explain what MCP is, when you need Snowflake's managed MCP server, and build a full working example on Snowflake. In Part 2: Building a Chat UI for Snowflake's Managed MCP Server, I build a small web console that talks to the servers from this post.TL;DR MCP (Model Context Protocol) is a contract for tools. A client like Claude, Cursor or ChatGPT asks a server "what tools do you have?" (tools/list) and then calls them (tools/call). That's most of it. Snowflake can host the MCP server for you. It's a schema-level object (CREATE MCP SERVER) that exposes Cortex Agents, Cortex Analyst, Cortex Search, read-only SQL and your own UDFs and procedures as tools. You don't run a container, and access is just Snowflake RBAC. Expose an agent, not raw SQL. Snowflake's own recommendation, and mine after testing: one Cortex Agent tool for business users, and a separate server with read-only SQL for engineers. Watch secondary roles with PATs. In my test, a role with no grant on a server could still list its tools, because the PAT had no role restriction and the user had DEFAULT_SECONDARY_ROLES = ALL (the default). Give PATs a ROLE_RESTRICTION, and keep OAUTH_USE_SECONDARY_ROLES at its default of NONE on OAuth integrations. Custom procedure tools must return one cell. A RETURNS TABLE procedure fails over MCP. Return a VARIANT instead. Auditing is free. Every MCP call shows up in query history under the real user and role, tagged with the MCP request ID. Everything below ran in a sandbox account on Oct 8, 2026. The outputs are real. The Slack message that started this Every Monday our ops lead used to post the same thing in the data channel: "Are the dashboards up to date? And can someone pull revenue by region for last quarter?" Someone on the data team would check when the loads last ran, write a query, paste a screenshot, and answer three follow-u

## CVE-2026-58497 - Uni-CLI: Legacy HTTP MCP transport accepted browser-originated localhost requests

DevFeed: [CVE-2026-58497 - Uni-CLI: Legacy HTTP MCP transport accepted browser-originated localhost requests](<https://devfeed.tech/articles/cve-2026-58497-uni-cli-legacy-http-mcp-transport-accepted-browser-originated-localhost-requests-98331.md>)

Original publisher: [Read original article](<https://cvefeed.io/vuln/detail/CVE-2026-58497>)

Author: CVEFEED.io; github.com

Published: 2026-10-23T15:17:34Z

Content type: news

Language: en

Sources: [Latest High/Critical Vulnerabilitiy Feed](<https://devfeed.tech/sources/latest-high-critical-vulnerabilitiy-feed.md>)

Topics: [Cross-origin resource sharing (CORS)](<https://devfeed.tech/topics/cors.md>), [frontend security](<https://devfeed.tech/topics/frontend-security.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [HTTP](<https://devfeed.tech/topics/http.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [browser](<https://devfeed.tech/tags/browser.md>), [cli](<https://devfeed.tech/tags/cli.md>), [cors](<https://devfeed.tech/tags/cors.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cve-2026-58497](<https://devfeed.tech/tags/cve-2026-58497.md>), [http](<https://devfeed.tech/tags/http.md>), [legacy](<https://devfeed.tech/tags/legacy.md>), [localhost](<https://devfeed.tech/tags/localhost.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [transport](<https://devfeed.tech/tags/transport.md>)

### AI overview

Before Uni-CLI 0.225.2, its legacy JSON-RPC-over-HTTP MCP transport accepted requests on loopback without checking browser Origin headers. A malicious webpage could send a simple POST to the local `/mcp` endpoint and, if the user had started the local MCP HTTP transport, invoke tools through the server. The Streamable HTTP transport already checked this browser-to-localhost boundary; the legacy stateless path did not. The issue is patched in 0.225.2 and does not change Uni-CLI's local-code-execution trust model.

### Source excerpt

CVE ID :CVE-2026-58497 Published : Oct. 23, 2026, 3:17 p.m. | 4 hours, 25 minutes ago Description :Uni-CLI gives people and agents one command interface for websites, browsers, desktop applications, local tools, and MCP. Prior to version 0.225.2, Uni-CLI exposed the legacy JSON-RPC-over-HTTP MCP transport on loopback without validating browser Origin headers before routing requests. A malicious web page could send a CORS simple POST request, such as text/plain, to the local /mcp endpoint and deliver a JSON-RPC body to the dispatcher. If the user had started the local MCP HTTP transport, that page could drive tools/call requests against the user's local Uni-CLI server. The Streamable HTTP transport already enforced this browser-to-localhost boundary. The legacy stateless HTTP path did not, so the two HTTP transports had drifted. This issue is about the browser-to-localhost boundary; it does not change Uni-CLI's local-code-execution trust model. This issue has been patched in version 0.225.2. Severity: 8.6 | HIGH Visit the link for more details, such as CVSS details, affected products, timeline, and more...

## Uber Automates Dependency Analysis Across Its Service Mesh

DevFeed: [Uber Automates Dependency Analysis Across Its Service Mesh](<https://devfeed.tech/articles/uber-automates-dependency-analysis-across-its-service-mesh-88464.md>)

Original publisher: [Read original article](<https://www.infoq.com/news/2026/10/uber-automated-dependency/>)

Author: Leela Kumili

Published: 2026-10-16T14:39:00Z

Content type: news

Language: en

Sources: [InfoQ](<https://devfeed.tech/sources/infoq.md>)

Topics: [service-mesh](<https://devfeed.tech/topics/service-mesh.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [OpenTelemetry](<https://devfeed.tech/topics/opentelemetry.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [architecture-design](<https://devfeed.tech/tags/architecture-design.md>), [dependency-management](<https://devfeed.tech/tags/dependency-management.md>), [development](<https://devfeed.tech/tags/development.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-tracing](<https://devfeed.tech/tags/distributed-tracing.md>), [failure](<https://devfeed.tech/tags/failure.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [microservices](<https://devfeed.tech/tags/microservices.md>), [middleware](<https://devfeed.tech/tags/middleware.md>), [middleware-management](<https://devfeed.tech/tags/middleware-management.md>), [news](<https://devfeed.tech/tags/news.md>), [observability](<https://devfeed.tech/tags/observability.md>), [opentelemetry](<https://devfeed.tech/tags/opentelemetry.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [service-mesh](<https://devfeed.tech/tags/service-mesh.md>), [service-reliability](<https://devfeed.tech/tags/service-reliability.md>), [uber-automated-dependency](<https://devfeed.tech/tags/uber-automated-dependency.md>)

### AI overview

Uber describes a middleware-based system that correlates caller and downstream service failures to classify dependencies as fail-close, fail-open, or unknown. It uses production metrics to map failure propagation across its service mesh, aiming to analyze rare failures faster than sampled tracing allows.

### Source excerpt

Uber has developed an automated dependency analysis system that uses RPC middleware and production metrics to correlate caller and downstream failures. The approach classifies dependencies as fail-close, fail-open, or unknown and addresses limitations of sampled distributed tracing for rare failures across Uber's large service mesh. By Leela Kumili

## Xorbits Xoscar: Unauthenticated RCE via Cloudpickle Deserialization in the Actor Pool

DevFeed: [Xorbits Xoscar: Unauthenticated RCE via Cloudpickle Deserialization in the Actor Pool](<https://devfeed.tech/articles/xorbits-xoscar-unauthenticated-rce-via-cloudpickle-deserialization-in-the-actor-pool-91621.md>)

Original publisher: [Read original article](<https://www.vulncheck.com/blog/xorbits-xoscar-unauth-cloudpickle-rce>)

Author: Valentin Lobstein (Chocapikk)

Published: 2026-10-16T00:00:00Z

Content type: news

Language: en

Sources: [VulnCheck](<https://devfeed.tech/sources/vulncheck.md>)

Topics: [Data serialization](<https://devfeed.tech/topics/data-serialization.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [Actor](<https://devfeed.tech/topics/actor.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [payload](<https://devfeed.tech/topics/payload.md>)

Tags: [authentication](<https://devfeed.tech/tags/authentication.md>), [code-execution](<https://devfeed.tech/tags/code-execution.md>), [no-authentication](<https://devfeed.tech/tags/no-authentication.md>), [rce](<https://devfeed.tech/tags/rce.md>), [remote-code-execution](<https://devfeed.tech/tags/remote-code-execution.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [unauthenticated-rce](<https://devfeed.tech/tags/unauthenticated-rce.md>), [xinference](<https://devfeed.tech/tags/xinference.md>)

### AI overview

CVE-2026-49950 enables unauthenticated code execution through Xoscar's actor-pool TCP channel: inbound messages are deserialized with `cloudpickle` before message type checks, with no channel authentication or restricted unpickler. A crafted body can run code as the pool process in one message; the reported lab exploit obtained root. All released Xoscar versions, including v0.11.2, are described as affected, with no fix available at the time of writing. Exposure depends on deployment: single-node pools bind locally on random ports, while Xinference distributed mode binds pools to `0.0.0.0` by default, making them reachable across the network. The article recommends keeping actor-pool ports on private segments and restricting access with network policy, firewall rules, or mutual authentication until a fix is released.

### Source excerpt

VulnCheck's Initial Access Intelligence team details an unauthenticated remote code execution in Xoscar, the actor and RPC framework behind the Xinference inference stack, where the actor pool deserializes every inbound TCP message with cloudpickle before it checks the message type and with no authentication.

## A2A agent cards

DevFeed: [A2A agent cards](<https://devfeed.tech/articles/a2a-agent-cards-89072.md>)

Original publisher: [Read original article](<https://specification.website/spec/agent-readiness/a2a-agent-cards/>)

Author: Joost de Valk

Published: 2026-10-15T00:00:00Z

Content type: tutorial

Language: en

Sources: [The Website Specification](<https://devfeed.tech/sources/the-website-specification.md>)

Topics: [A2A protocol](<https://devfeed.tech/topics/a2a-protocol.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [gRPC](<https://devfeed.tech/topics/grpc.md>)

Tags: [a2a-protocol](<https://devfeed.tech/tags/a2a-protocol.md>), [agent-readiness](<https://devfeed.tech/tags/agent-readiness.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [json](<https://devfeed.tech/tags/json.md>), [json-rpc](<https://devfeed.tech/tags/json-rpc.md>), [optional](<https://devfeed.tech/tags/optional.md>), [rpc](<https://devfeed.tech/tags/rpc.md>)

### AI overview

A2A agent cards are JSON manifests served at `/.well-known/agent-card.json` that let agents discover another agent's skills, capabilities, and endpoints. The article explains the required and optional card fields, transport choices, publishing and compatibility guidance, common mistakes, and validation steps for implementing a card aligned with A2A 1.0.

### Source excerpt

The Agent-to-Agent (A2A) protocol lets an autonomous agent find another autonomous agent and call it over JSON-RPC. Discovery hinges on a single well-known file: `/.well-known/agent-card.json`. Relevant whenever your service exposes agentic behaviour another agent might want to delegate to.

## Figma MCP 403 Forbidden: Why client\_name allowlists fail

DevFeed: [Figma MCP 403 Forbidden: Why client\_name allowlists fail](<https://devfeed.tech/articles/figma-mcp-403-forbidden-why-client-name-allowlists-fail-65068.md>)

Original publisher: [Read original article](<https://workos.com/blog/figma-mcp-403-forbidden-client-name>)

Author: WorkOS

Published: 2026-10-05T00:00:00Z

Content type: article

Language: en

Sources: [WorkOS](<https://devfeed.tech/sources/workos-blog.md>)

Topics: [MSP MCP](<https://devfeed.tech/topics/msp-mcp.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Figma AI](<https://devfeed.tech/topics/figma-ai.md>)

Tags: [403](<https://devfeed.tech/tags/403.md>), [access-control](<https://devfeed.tech/tags/access-control.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [bedrock-agentcore](<https://devfeed.tech/tags/bedrock-agentcore.md>), [claude](<https://devfeed.tech/tags/claude.md>), [cli](<https://devfeed.tech/tags/cli.md>), [client](<https://devfeed.tech/tags/client.md>), [code](<https://devfeed.tech/tags/code.md>), [codex](<https://devfeed.tech/tags/codex.md>), [copilot](<https://devfeed.tech/tags/copilot.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-server](<https://devfeed.tech/tags/mcp-server.md>), [remote-mcp-server](<https://devfeed.tech/tags/remote-mcp-server.md>)

### AI overview

Figma's remote MCP server uses the self-asserted client_name in dynamic client registration to allowlist clients, so changing a client name can bypass the check while supported tools may still be denied. The article explains how Client ID Metadata Documents improve identity claims but cannot verify native CLI or desktop binaries using localhost redirects. It recommends domain checks for hosted clients, scope-based access, customer-admin policies, clearer onboarding, and audit logs.

### Source excerpt

igma's remote MCP server admits clients by the client_name they send during dynamic client registration. Why that check proves nothing, what CIMD fixes and what it doesn't, and what to gate on instead.

## Cato VPN Client: Split-Tunnel and Privilege Escalation (CVE-2026-10739)

DevFeed: [Cato VPN Client: Split-Tunnel and Privilege Escalation (CVE-2026-10739)](<https://devfeed.tech/articles/cato-vpn-client-split-tunnel-and-privilege-escalation-cve-2026-10739-92085.md>)

Original publisher: [Read original article](<http://blog.quarkslab.com/cato-vpn-client-split-tunnel-and-privilege-escalation-cve-2026-10739.html>)

Author: Lucas Laise

Published: 2026-09-30T22:00:00Z

Content type: article

Language: en

Sources: [Quarkslab](<https://devfeed.tech/sources/quarkslab.md>)

Topics: [Virtual Private Network](<https://devfeed.tech/topics/vpn.md>), [.NET Framework](<https://devfeed.tech/topics/net-framework.md>), [out-of-process](<https://devfeed.tech/topics/out-of-process.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [cve remediation](<https://devfeed.tech/topics/cve-remediation.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [cato-networks](<https://devfeed.tech/tags/cato-networks.md>), [client](<https://devfeed.tech/tags/client.md>), [cve](<https://devfeed.tech/tags/cve.md>), [cve-2026-10739](<https://devfeed.tech/tags/cve-2026-10739.md>), [delete](<https://devfeed.tech/tags/delete.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [flow](<https://devfeed.tech/tags/flow.md>), [installer](<https://devfeed.tech/tags/installer.md>), [local](<https://devfeed.tech/tags/local.md>), [local-privilege-escalation](<https://devfeed.tech/tags/local-privilege-escalation.md>), [lpe](<https://devfeed.tech/tags/lpe.md>), [named-pipe](<https://devfeed.tech/tags/named-pipe.md>), [net](<https://devfeed.tech/tags/net.md>), [pentest](<https://devfeed.tech/tags/pentest.md>), [privilege-escalation](<https://devfeed.tech/tags/privilege-escalation.md>), [rollback](<https://devfeed.tech/tags/rollback.md>), [split](<https://devfeed.tech/tags/split.md>), [system](<https://devfeed.tech/tags/system.md>), [upload](<https://devfeed.tech/tags/upload.md>), [vpn](<https://devfeed.tech/tags/vpn.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>), [windows](<https://devfeed.tech/tags/windows.md>)

### AI overview

Cato VPN Client has a local privilege escalation on Windows, confirmed on versions 6.2.0 and 6.4.6; the vendor says versions before 6.12.6 are affected. The split-tunnel upload service accepts a client-supplied SID over a named pipe and uses it in a file path without rejecting traversal components. An attacker can pass the pipe's certificate check by injecting a payload into signed CatoClient.exe, then submit invalid split-tunnel data to trigger SYSTEM cleanup of a traversed path. The writeup redirects that delete to C:\Config.Msi and uses a Windows Installer rollback chain to obtain SYSTEM code execution. The chain depends on the signed-process origin, crafted protobuf messages, invalid-file cleanup, and filesystem redirection.

### Source excerpt

Local privilege escalation in the Cato VPN Client through the split-tunnel upload flow, from named pipe to SYSTEM delete and Windows Installer rollback.

## Beyond Kubernetes at Modal: How to Scale 1 Million Concurrent Sandboxes in Seconds

DevFeed: [Beyond Kubernetes at Modal: How to Scale 1 Million Concurrent Sandboxes in Seconds](<https://devfeed.tech/articles/beyond-kubernetes-at-modal-how-to-scale-1-million-concurrent-sandboxes-in-seconds-58939.md>)

Original publisher: [Read original article](<https://www.infoq.com/news/2026/09/modal-scaling-sandboxes/>)

Author: Sergio De Simone

Published: 2026-09-23T14:00:00Z

Content type: news

Language: en

Sources: [InfoQ](<https://devfeed.tech/sources/infoq.md>)

Topics: [Kubernetes](<https://devfeed.tech/topics/kubernetes.md>), [systems](<https://devfeed.tech/topics/systems.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>), [Redis](<https://devfeed.tech/topics/redis.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [etcd](<https://devfeed.tech/topics/etcd.md>), [scheduling](<https://devfeed.tech/topics/scheduling.md>), [container](<https://devfeed.tech/topics/container.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-ml-data-engineering](<https://devfeed.tech/tags/ai-ml-data-engineering.md>), [architecture-design](<https://devfeed.tech/tags/architecture-design.md>), [balancing](<https://devfeed.tech/tags/balancing.md>), [containers](<https://devfeed.tech/tags/containers.md>), [development](<https://devfeed.tech/tags/development.md>), [devops](<https://devfeed.tech/tags/devops.md>), [etcd](<https://devfeed.tech/tags/etcd.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [ml-data-engineering](<https://devfeed.tech/tags/ml-data-engineering.md>), [modal-scaling-sandboxes](<https://devfeed.tech/tags/modal-scaling-sandboxes.md>), [news](<https://devfeed.tech/tags/news.md>), [performance-scalability](<https://devfeed.tech/tags/performance-scalability.md>), [redis](<https://devfeed.tech/tags/redis.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [sandboxes](<https://devfeed.tech/tags/sandboxes.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

This news report describes how Modal redesigned its sandbox infrastructure to support millions of concurrent sandboxes and tens of thousands of creations per second. The architecture replaces centralized coordination with horizontally scalable scheduling servers, worker-local state, direct RPC requests, and a Redis stream for worker state.

### Source excerpt

In a recent article, Colin Weld and Connor Adams, staff engineers at Modal, describe how they rebuilt their sandbox infrastructure from the ground up to support millions of concurrent sandboxes and tens of thousands of sandbox creations per second. By Sergio De Simone

## Building Composite Model Context Protocol (MCP) Gateways in TypeScript

DevFeed: [Building Composite Model Context Protocol (MCP) Gateways in TypeScript](<https://devfeed.tech/articles/building-composite-model-context-protocol-mcp-gateways-in-typescript-55834.md>)

Original publisher: [Read original article](<https://www.sitepoint.com/building-composite-model-context-protocol-mcp-gateways-in-typescript/>)

Author: SitePoint Team

Published: 2026-09-21T14:38:42Z

Content type: tutorial

Language: en

Sources: [SitePoint](<https://devfeed.tech/sources/sitepoint.md>)

Topics: [Model Context Protocol](<https://devfeed.tech/topics/model-context-protocol.md>), [gateway](<https://devfeed.tech/topics/gateway.md>), [TypeScript](<https://devfeed.tech/topics/typescript.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>), [Authentication](<https://devfeed.tech/topics/authentication.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [apis](<https://devfeed.tech/tags/apis.md>), [auth](<https://devfeed.tech/tags/auth.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [json-rpc](<https://devfeed.tech/tags/json-rpc.md>), [limits](<https://devfeed.tech/tags/limits.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-gateway](<https://devfeed.tech/tags/mcp-gateway.md>), [programming](<https://devfeed.tech/tags/programming.md>), [rag](<https://devfeed.tech/tags/rag.md>), [typescript](<https://devfeed.tech/tags/typescript.md>)

### AI overview

A tutorial on building a composite Model Context Protocol gateway in TypeScript. It explains how to aggregate multiple upstream MCP servers, resolve tool-name collisions, route JSON-RPC requests, enforce per-tool authorization, and apply token-bucket rate limiting.

### Source excerpt

Build a unified Model Context Protocol (MCP) gateway in TypeScript that aggregates upstream MCP servers, resolves tool collisions, and handles auth and rate limits. Continue reading Building Composite Model Context Protocol (MCP) Gateways in TypeScript on SitePoint.

## MCP goes stateless, and Quarkus already implements it

DevFeed: [MCP goes stateless, and Quarkus already implements it](<https://devfeed.tech/articles/mcp-goes-stateless-and-quarkus-already-implements-it-58518.md>)

Original publisher: [Read original article](<https://quarkus.io/blog/mcp-stateless/>)

Author: Kevin Dubois (https://twitter.com/kevindubois)

Published: 2026-09-21T00:00:00Z

Content type: article

Language: en

Sources: [Quarkus - Supersonic Subatomic Java](<https://devfeed.tech/sources/quarkus-supersonic-subatomic-java.md>)

Topics: [Model Context Protocol](<https://devfeed.tech/topics/model-context-protocol.md>), [Quarkus](<https://devfeed.tech/topics/quarkus.md>), [stateless](<https://devfeed.tech/topics/stateless.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [HTTP](<https://devfeed.tech/topics/http.md>), [JSON](<https://devfeed.tech/topics/json.md>), [sessions](<https://devfeed.tech/topics/sessions.md>), [versioning](<https://devfeed.tech/topics/versioning.md>), [client](<https://devfeed.tech/topics/client.md>), [Server](<https://devfeed.tech/topics/server.md>)

Tags: [http](<https://devfeed.tech/tags/http.md>), [json-rpc](<https://devfeed.tech/tags/json-rpc.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [quarkus](<https://devfeed.tech/tags/quarkus.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [sessions](<https://devfeed.tech/tags/sessions.md>), [stateless](<https://devfeed.tech/tags/stateless.md>), [versioning](<https://devfeed.tech/tags/versioning.md>)

### AI overview

This article explains how the July 28, 2026 revision of the Model Context Protocol removes stateful session handshakes in favor of self-contained JSON-RPC requests. It describes how Quarkus MCP Server 2.0.x supports the change, including metadata, version negotiation, capability discovery, and multi-round-trip requests.

### Source excerpt

If you've been running Model Context Protocol (MCP) servers, you might have bumped into the limitations of stateful sessions. Until recently, a client had to perform an initialize handshake, get a session ID, and stick to that specific server instance for the duration of the connection. For a single-instance setup, this is fine. But the moment you move to a distributed environment, you're forced into using sticky sessions or a shared session store just to keep the connection alive. It's an unnecessary hurdle for what should essentially be a set of HTTP calls. The 2026-07-28 revision of the MCP specification finally solves this by going stateless, and the 2.0.x line of the Quarkus MCP Server already supports these changes out of the box. Here is a look at how the protocol has evolved in this latest version and how to use it in Quarkus. Moving from sessions to self-contained requests The core shift in the July 28 revision is the removal of the stateful session handshake. Instead of negotiating a session once, every request is now a self-contained JSON-RPC message. Everything the server needs -- protocol version, client identity, and capabilities -- is now packed into a _meta field within the request. If you're using Streamable HTTP, this metadata is mirrored in headers (MCP-Protocol-Version, Mcp-Method, and Mcp-Name) so the server can route or reject the request before it even parses the body. Since there is no session state to track, any server instance can handle any request. An MCP server is now as easy to scale horizontally as any other stateless REST API. What about versioning? Negotiation has also moved to the request level. If a server doesn't recognize the version requested in a specific call, it responds with an UnsupportedProtocolVersionError and lists its supported versions. For clients that want to be proactive, there is now a server/discover RPC to check capabilities upfront. Multi Round-Trip Requests Going stateless creates one obvious problem though: how

## Bringing QUIC to Seastar

DevFeed: [Bringing QUIC to Seastar](<https://devfeed.tech/articles/bringing-quic-to-seastar-17377.md>)

Original publisher: [Read original article](<https://www.scylladb.com/2026/09/14/bringing-quic-to-seastar/>)

Author: Cynthia Dunlop

Published: 2026-09-14T13:02:07Z

Content type: article

Language: en

Sources: [ScyllaDB](<https://devfeed.tech/sources/scylladb.md>)

Topics: [Seastar](<https://devfeed.tech/topics/seastar.md>), [Network](<https://devfeed.tech/topics/network.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [C++](<https://devfeed.tech/topics/c-plus-plus.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [Protocol (disambiguation)](<https://devfeed.tech/topics/protocol.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [c-plus-plus](<https://devfeed.tech/tags/c-plus-plus.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [latency](<https://devfeed.tech/tags/latency.md>), [network](<https://devfeed.tech/tags/network.md>), [networking](<https://devfeed.tech/tags/networking.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [scylladb](<https://devfeed.tech/tags/scylladb.md>), [seastar](<https://devfeed.tech/tags/seastar.md>), [udp](<https://devfeed.tech/tags/udp.md>)

### AI overview

This article describes a University of Warsaw student project conducted with ScyllaDB to implement a QUIC transport for Seastar using the sans-I/O ngtcp2 library. It also adapts Seastar's RPC layer to operate over QUIC and reports benchmark results showing predictable overhead on a lossless loopback and benefits when packets are dropped.

### Source excerpt

We built a QUIC transport for Seastar on top of ngtcp2's sans-I/O state machine, then adapted RPC to it twice: 1) as a one-to-one socket replacement, and 2) a QUIC-aware approach that opens a fresh stream per call.

## Solana Changelog: September 3, 2026

DevFeed: [Solana Changelog: September 3, 2026](<https://devfeed.tech/articles/solana-changelog-september-3-2026-42719.md>)

Original publisher: [Read original article](<https://solana.com/news/solana-changelog-september-3-2026>)

Author: Solana Foundation

Published: 2026-09-10T20:16:00Z

Content type: news

Language: en

Sources: [Solana News Feed](<https://devfeed.tech/sources/solana-news-feed.md>)

Topics: [Solana](<https://devfeed.tech/topics/solana.md>), [changelog](<https://devfeed.tech/topics/changelog.md>), [SDK](<https://devfeed.tech/topics/sdk.md>), [releases](<https://devfeed.tech/topics/releases.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Network](<https://devfeed.tech/topics/network.md>), [Arm](<https://devfeed.tech/topics/arm.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [arm](<https://devfeed.tech/tags/arm.md>), [blockchain](<https://devfeed.tech/tags/blockchain.md>), [blockchain-technology](<https://devfeed.tech/tags/blockchain-technology.md>), [changelog](<https://devfeed.tech/tags/changelog.md>), [crypto-news](<https://devfeed.tech/tags/crypto-news.md>), [cryptocurrency](<https://devfeed.tech/tags/cryptocurrency.md>), [defi](<https://devfeed.tech/tags/defi.md>), [developers](<https://devfeed.tech/tags/developers.md>), [js-sdk](<https://devfeed.tech/tags/js-sdk.md>), [network](<https://devfeed.tech/tags/network.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [nfts](<https://devfeed.tech/tags/nfts.md>), [podcasts](<https://devfeed.tech/tags/podcasts.md>), [releases](<https://devfeed.tech/tags/releases.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [sdk](<https://devfeed.tech/tags/sdk.md>), [solana](<https://devfeed.tech/tags/solana.md>), [solana-ecosystem](<https://devfeed.tech/tags/solana-ecosystem.md>), [technology](<https://devfeed.tech/tags/technology.md>), [web3](<https://devfeed.tech/tags/web3.md>)

### AI overview

A weekly Solana engineering newsletter covering new Agave, Web3.js, Solana Kit, and program SDK versions, runtime bytecode verification discussions, validator-client changes, transaction ordering, address restrictions, and shred-recovery work.

### Source excerpt

Firedancer adopts three RPC calls and advances ARM support, while three mainnet feature gates and new Agave, Web3.js, and Solana Kit versions land.

## MCP vs. API: Key Differences and When To Use Each

DevFeed: [MCP vs. API: Key Differences and When To Use Each](<https://devfeed.tech/articles/mcp-vs-api-key-differences-and-when-to-use-each-51531.md>)

Original publisher: [Read original article](<https://blog.n8n.io/mcp-vs-api/>)

Author: n8n team

Published: 2026-09-10T16:41:50Z

Content type: tutorial

Language: en

Sources: [n8n](<https://devfeed.tech/sources/n8n-blog.md>)

Topics: [API](<https://devfeed.tech/topics/api.md>), [Model Context Protocol](<https://devfeed.tech/topics/model-context-protocol.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [JSON](<https://devfeed.tech/topics/json.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [OpenAPI Specification](<https://devfeed.tech/topics/openapi.md>), [n8n](<https://devfeed.tech/topics/n8n.md>), [OAuth](<https://devfeed.tech/topics/oauth.md>), [GraphQL](<https://devfeed.tech/topics/graphql.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [api](<https://devfeed.tech/tags/api.md>), [graphql](<https://devfeed.tech/tags/graphql.md>), [guides](<https://devfeed.tech/tags/guides.md>), [json](<https://devfeed.tech/tags/json.md>), [json-rpc](<https://devfeed.tech/tags/json-rpc.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-vs-api](<https://devfeed.tech/tags/mcp-vs-api.md>), [model-context-protocol](<https://devfeed.tech/tags/model-context-protocol.md>), [n8n](<https://devfeed.tech/tags/n8n.md>), [oauth](<https://devfeed.tech/tags/oauth.md>), [openapi](<https://devfeed.tech/tags/openapi.md>)

### AI overview

This article compares traditional APIs with the Model Context Protocol (MCP), explaining how their interfaces, consumers, capability discovery, and state handling differ. It also describes how MCP connects AI models to external tools through discoverable schemas and JSON-RPC, and when systems may use MCP, APIs, or both.

### Source excerpt

MCP vs. API: Learn how the Model Context Protocol differs from traditional APIs, when to use each, and how to combine both in n8n.

## 🍔🧠 How OpenAI Built GPT-Live for Low-Latency Voice AI

DevFeed: [🍔🧠 How OpenAI Built GPT-Live for Low-Latency Voice AI](<https://devfeed.tech/articles/how-openai-built-gpt-live-for-low-latency-voice-ai-18127.md>)

Original publisher: [Read original article](<https://hungrymindsdev.substack.com/p/how-openai-built-gpt-live-for-low>)

Author: Alexandre Zajac

Published: 2026-09-07T15:31:33Z

Content type: article

Language: en

Sources: [Hungry Minds](<https://devfeed.tech/sources/hungry-minds.md>)

Topics: [Conversational AI](<https://devfeed.tech/topics/conversational-ai.md>), [Latency](<https://devfeed.tech/topics/latency.md>), [voice ai](<https://devfeed.tech/topics/voice-ai.md>), [OpenAI](<https://devfeed.tech/topics/openai.md>), [real-time](<https://devfeed.tech/topics/real-time.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Go Language](<https://devfeed.tech/topics/go-language.md>)

Tags: [go](<https://devfeed.tech/tags/go.md>), [low-latency](<https://devfeed.tech/tags/low-latency.md>), [openai](<https://devfeed.tech/tags/openai.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [voice-ai](<https://devfeed.tech/tags/voice-ai.md>)

### AI overview

The article explains how OpenAI built GPT-Live as a low-latency, full-duplex voice system. It describes a fast audio path that listens and speaks simultaneously, asynchronous delegation of deeper reasoning and backend work, stateful handoffs, optimized protocol boundaries, and a Go-based media frontend.

### Source excerpt

PLUS: Zero-knowledge proofs ⚡, Design.md agent automation 👨💻, System Design Docs 101 📚

## MCP Logging in Spring AI

DevFeed: [MCP Logging in Spring AI](<https://devfeed.tech/articles/mcp-logging-in-spring-ai-4502.md>)

Original publisher: [Read original article](<https://www.baeldung.com/spring-ai-mcp-logging>)

Author: Burak Gökmen

Published: 2026-09-06T23:19:21Z

Content type: tutorial

Language: en

Sources: [Baeldung](<https://devfeed.tech/sources/baeldung.md>)

Topics: [Model Context Protocol](<https://devfeed.tech/topics/model-context-protocol.md>), [log management](<https://devfeed.tech/topics/log-management.md>), [AI Development](<https://devfeed.tech/topics/ai-development.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [telemetry](<https://devfeed.tech/topics/telemetry.md>)

Tags: [logging](<https://devfeed.tech/tags/logging.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-server](<https://devfeed.tech/tags/mcp-server.md>), [opentelemetry](<https://devfeed.tech/tags/opentelemetry.md>), [popular](<https://devfeed.tech/tags/popular.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [spring](<https://devfeed.tech/tags/spring.md>), [spring-ai](<https://devfeed.tech/tags/spring-ai.md>), [spring-ai-mcp-popular](<https://devfeed.tech/tags/spring-ai-mcp-popular.md>), [spring-boot](<https://devfeed.tech/tags/spring-boot.md>), [tutorial](<https://devfeed.tech/tags/tutorial.md>)

### AI overview

A tutorial on MCP logging in Spring AI, covering server-to-client log notifications, client-controlled severity levels, and migration guidance following the feature's deprecation.

### Source excerpt

Learn about MCP logging in Spring AI, both in an MCP server and a client. The post MCP Logging in Spring AI first appeared on Baeldung.

## Kubernetes v1.37: etcd RangeStream Cuts Memory Use on Large List Reads

DevFeed: [Kubernetes v1.37: etcd RangeStream Cuts Memory Use on Large List Reads](<https://devfeed.tech/articles/kubernetes-v1-37-etcd-rangestream-cuts-memory-use-on-large-list-reads-4576.md>)

Original publisher: [Read original article](<https://kubernetes.io/blog/2026/09/01/kubernetes-v1-37-etcd-range-stream/>)

Author: Jeffrey Ying

Published: 2026-09-01T18:30:00Z

Content type: release

Language: en

Sources: [Kubernetes Blog](<https://devfeed.tech/sources/kubernetes-blog.md>)

Topics: [Kubernetes](<https://devfeed.tech/topics/kubernetes.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>), [Cache](<https://devfeed.tech/topics/cache.md>)

Tags: [cache](<https://devfeed.tech/tags/cache.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [memory](<https://devfeed.tech/tags/memory.md>), [releases](<https://devfeed.tech/tags/releases.md>), [rpc](<https://devfeed.tech/tags/rpc.md>), [streaming](<https://devfeed.tech/tags/streaming.md>)

### AI overview

Kubernetes v1.37 promotes etcd RangeStream to beta. With etcd v3.7, streamed collection reads reduce and stabilize memory use by processing chunks instead of assembling entire pages.

### Source excerpt

I am excited to announce that etcd RangeStream is graduating to beta in Kubernetes v1.37. Paired with etcd v3.7, it reduces the memory the API server and etcd need to read a large collection, and makes peak usage more predictable. The cost of large reads The API server serves most list and watch requests from its in-memory watch cache. Populating that cache requires reading a resource's full state from etcd, at startup and on every re-initialization. For a resource with many objects, or large ones, such as Pods, that read is expensive. The API server already paginated these reads, asking etcd for a fixed number of keys at a time rather than the whole collection at once. But a page bounded by key count has no awareness of object size, so a page of large objects can still be very large. That makes memory usage hard to predict, and a bad combination of object size and concurrent reads can be enough to trigger an OOM. etcd's unary Range assembles each page in full before sending it, and the API server holds it while decoding, so the same payload sits in memory on both sides at once. Most of that cost lands on etcd, which is also where streaming helps most. Streaming reads with RangeStream etcd v3.7 adds a streaming version of that read, the RangeStream RPC. It takes the same RangeRequest as Range and returns the same result set, but instead of building the whole response up front, etcd splits it into chunks and streams them. Chunk size is tuned adaptively to the values being returned, so a collection of large objects is bounded by bytes rather than by a key count, and memory is freed as the stream progresses instead of being held until a whole page is assembled. When the feature is enabled, the API server uses RangeStream wherever it reads a whole collection out of etcd. This includes watch cache initialization, and the fallback paths where a list request cannot be served from the cache and reads etcd directly. In either case the API server decodes each chunk as it arri

[Next page](<https://devfeed.tech/topics/rpc.md?cursor=WyIyMDI2LTA5LTAxVDE4OjMwOjAwKzAwOjAwIiwgImEwMWE0YTRiLTllZDItNDhlZi1hN2Q1LTljN2MxYTk0MDZhNSJd>)