# TCP/IP

TCP/IP is a suite of communications protocols, specifically Transmission Control Protocol and Internet Protocol, that forms a technical foundation of the Internet.

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

## Could pigeons deliver the Internet?

DevFeed: [Could pigeons deliver the Internet?](<https://devfeed.tech/articles/could-pigeons-deliver-the-internet-131302.md>)

Original publisher: [Read original article](<https://blog.apnic.net/2026/11/13/could-pigeons-deliver-the-internet/>)

Author: Tijana Milentijevic

Published: 2026-11-12T23:37:47Z

Content type: article

Language: en

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

Topics: [Internet](<https://devfeed.tech/topics/internet.md>), [communications](<https://devfeed.tech/topics/communications.md>), [Internet Engineering Task Force (IETF)](<https://devfeed.tech/topics/ietf.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-training](<https://devfeed.tech/tags/ai-training.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [applications](<https://devfeed.tech/tags/applications.md>), [april-fools](<https://devfeed.tech/tags/april-fools.md>), [area](<https://devfeed.tech/tags/area.md>), [based](<https://devfeed.tech/tags/based.md>), [birds](<https://devfeed.tech/tags/birds.md>), [cables](<https://devfeed.tech/tags/cables.md>), [carrier](<https://devfeed.tech/tags/carrier.md>), [guest-post](<https://devfeed.tech/tags/guest-post.md>), [internet](<https://devfeed.tech/tags/internet.md>), [measurement](<https://devfeed.tech/tags/measurement.md>), [need](<https://devfeed.tech/tags/need.md>), [network](<https://devfeed.tech/tags/network.md>), [post](<https://devfeed.tech/tags/post.md>), [tech-matters](<https://devfeed.tech/tags/tech-matters.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The paper models pigeon delivery as a directed communication network and minimizes the number of birds, which fly only to their home location and cannot be reused for another delivery. Direct single-hop delivery needs one pigeon per source-destination pair; for 20 communicating locations, that is 380. Allowing messages to stop and be combined reduces bird use but makes the two-hop and multihop optimization problems NP-hard. The polynomial-time Coordinator algorithm gathers messages at the location with the most links, then distributes them by destination; it uses 38 pigeons in the 20-location example and is guaranteed to use at most twice the optimum. Integer linear programs can find exact optima for small two-hop and multihop instances. The model assumes unlimited carrying capacity and known demands, and counts pigeons rather than delay or transport energy. The authors connect the optimization approach to predictable data-centre traffic and vehicle or commodity routing.

### Source excerpt

Guest Post: What happens when we replace network cables with carrier pigeons? We worked out how many birds you would need.

## USBnet: Internet via USB (and HID!)

DevFeed: [USBnet: Internet via USB (and HID!)](<https://devfeed.tech/articles/usbnet-internet-via-usb-and-hid-130701.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/12/usbnet-internet-via-usb-and-hid/>)

Author: Anne Barela

Published: 2026-11-12T18:00:58Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [USB](<https://devfeed.tech/topics/usb.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Internet](<https://devfeed.tech/topics/internet.md>), [proxy](<https://devfeed.tech/topics/proxy.md>)

Tags: [allowed](<https://devfeed.tech/tags/allowed.md>), [block](<https://devfeed.tech/tags/block.md>), [board](<https://devfeed.tech/tags/board.md>), [button](<https://devfeed.tech/tags/button.md>), [ch32v467](<https://devfeed.tech/tags/ch32v467.md>), [classes](<https://devfeed.tech/tags/classes.md>), [client](<https://devfeed.tech/tags/client.md>), [computers](<https://devfeed.tech/tags/computers.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [github](<https://devfeed.tech/tags/github.md>), [hid](<https://devfeed.tech/tags/hid.md>), [http](<https://devfeed.tech/tags/http.md>), [http-proxy](<https://devfeed.tech/tags/http-proxy.md>), [internet](<https://devfeed.tech/tags/internet.md>), [local](<https://devfeed.tech/tags/local.md>), [microcontrollers](<https://devfeed.tech/tags/microcontrollers.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [proxy](<https://devfeed.tech/tags/proxy.md>), [risc-v](<https://devfeed.tech/tags/risc-v.md>), [server](<https://devfeed.tech/tags/server.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [upstream](<https://devfeed.tech/tags/upstream.md>), [usb](<https://devfeed.tech/tags/usb.md>), [usbnet](<https://devfeed.tech/tags/usbnet.md>)

### AI overview

USBnet tunnels TCP connections between two computers through a CH32V467WEU6 board connected to both over USB. Its client exposes an HTTP proxy at 127.0.0.1:8080, and its server opens upstream connections. To work around computers that restrict USB device classes, it can carry packets over HID and present as a mouse. Holding USER while resetting enters configuration mode, where the board appears as USBnet Configuration (1a86:fe53) and its flash-stored profile can be edited; port modes are relay and custom. Performance figures have different conditions: relay measurements were over a local link and mainly reflect the tunnel, while emulated figures used speedtest.net over the internet and include the connection. Firmware uses WCH's RISC-V toolchain from MounRiver.

### Source excerpt

Ismael has published USBnet on GitHub, a project that tunnels TCP connections between two computers over a single CH32V467WEU6 board. The board plugs into both machines over USB. On one side the client offers a local HTTP proxy at 127.0.0.1:8080, and on the other side the server opens the upstream TCP connections to the internet. [...]

## The DOs and don'ts of post-quantum DNSSEC

DevFeed: [The DOs and don'ts of post-quantum DNSSEC](<https://devfeed.tech/articles/the-dos-and-don-ts-of-post-quantum-dnssec-123783.md>)

Original publisher: [Read original article](<https://blog.apnic.net/2026/11/09/the-dos-and-donts-of-post-quantum-dnssec/>)

Author: Geoff Huston

Published: 2026-11-09T05:34:30Z

Content type: article

Language: en

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

Topics: [DNSSEC](<https://devfeed.tech/topics/dnssec.md>), [Post-quantum cryptography](<https://devfeed.tech/topics/post-quantum-cryptography.md>), [Domain Name System (DNS)](<https://devfeed.tech/topics/dns-resolution-process.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [Quantum Computing](<https://devfeed.tech/topics/quantum-computing.md>)

Tags: [dns](<https://devfeed.tech/tags/dns.md>), [dnssec](<https://devfeed.tech/tags/dnssec.md>), [post-quantum](<https://devfeed.tech/tags/post-quantum.md>), [post-quantum-cryptography](<https://devfeed.tech/tags/post-quantum-cryptography.md>), [quantum](<https://devfeed.tech/tags/quantum.md>), [queries](<https://devfeed.tech/tags/queries.md>), [resolvers](<https://devfeed.tech/tags/resolvers.md>), [responses](<https://devfeed.tech/tags/responses.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [tech-matters](<https://devfeed.tech/tags/tech-matters.md>), [udp](<https://devfeed.tech/tags/udp.md>)

### AI overview

The example shows ML-DSA-44 DNSSEC responses can exceed the recommended UDP payload size: one measured response was 2,551 bytes versus 191 bytes with ECDSA P-256, triggering UDP truncation and a TCP retry. The text explains that IP fragmentation is unreliable, while TCP is already part of DNS, and that the DO bit requests DNSSEC records rather than validation. An Internet draft proposes that non-validating resolvers clear DO and cache responses by DO setting, fetching signatures only when later requested; this could reduce unnecessary large responses and TCP load but adds resolver complexity and queries. The authors present the proposal as an unresolved idea. Measurements cited show 60% of users' queries set DO, while 48% are behind resolvers expected to validate; caching and the small share of queried names in signed zones temper the infrastructure impact.

### Source excerpt

Post-quantum signatures could dramatically enlarge DNSSEC responses, shifting many queries from UDP to TCP. Should resolvers limit unnecessary DNSSEC data, or is greater TCP use inevitable?

## Global Surveillance Under the Lens: Reconstructing Passive Dolev-Yao Intruders

DevFeed: [Global Surveillance Under the Lens: Reconstructing Passive Dolev-Yao Intruders](<https://devfeed.tech/articles/global-surveillance-under-the-lens-reconstructing-passive-dolev-yao-intruders-119907.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.07170>)

Author: Paolo Modesti

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Internet Traffic](<https://devfeed.tech/topics/internet-traffic.md>), [network-intelligence](<https://devfeed.tech/topics/network-intelligence.md>), [test](<https://devfeed.tech/topics/test.md>), [Automated reasoning](<https://devfeed.tech/topics/automated-reasoning.md>)

Tags: [automated](<https://devfeed.tech/tags/automated.md>), [capture](<https://devfeed.tech/tags/capture.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [channel](<https://devfeed.tech/tags/channel.md>), [compiler](<https://devfeed.tech/tags/compiler.md>), [concurrent](<https://devfeed.tech/tags/concurrent.md>), [cryptographic](<https://devfeed.tech/tags/cryptographic.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [defensive](<https://devfeed.tech/tags/defensive.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [execution](<https://devfeed.tech/tags/execution.md>), [formal](<https://devfeed.tech/tags/formal.md>), [framework](<https://devfeed.tech/tags/framework.md>), [intruder](<https://devfeed.tech/tags/intruder.md>), [java](<https://devfeed.tech/tags/java.md>), [jdk](<https://devfeed.tech/tags/jdk.md>), [mapping](<https://devfeed.tech/tags/mapping.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [methodology](<https://devfeed.tech/tags/methodology.md>), [multiple](<https://devfeed.tech/tags/multiple.md>), [network](<https://devfeed.tech/tags/network.md>)

### AI overview

The framework reconstructs passive Dolev-Yao intruder executions from PCAP traffic and checks them against AnBx protocol specifications. Given a capture, specification and deployment metadata, it extracts Java-serialized objects from TCP streams, assigns channels to protocol steps and sessions, then runs compiler-generated semantic checks. It addresses multiple objects and references in streams, ambiguous mappings of interleaved connections to concurrent sessions, and validation of candidate reconstructions. Across 121 AnBx test-suite protocols, the standard JDK cryptographic provider recovered 97.8% of expected objects and achieved 94.2% successful reconstruction for single sessions; success required complete object recovery and a clean intruder execution. For 112 parallel-capable protocols, two- and three-session reconstructions each succeeded 89.3% of the time, with about 97.2% object recovery. Tests with role mismatches and mutated traces showed the checks detected semantic inconsistencies.

### Source excerpt

arXiv:2611.07170v1 Announce Type: new Abstract: Global surveillance systems record large volumes of network traffic for retrospective analysis, but packet captures alone reveal little about protocol-level semantics. This paper presents an automated, specification-guided methodology for reconstructing passive Dolev-Yao intruder executions from captured network traffic and validating the reconstructed executions against a formal protocol specification. The framework takes a PCAP capture, an AnBx specification, and the associated deployment metadata; it extracts Java-serialised objects from TCP streams, reconstructs protocol sessions and channel-to-step assignments, and generates a passive intruder that evaluates the Dolev-Yao derived checks produced by the AnBx compiler. Three reconstruction challenges are addressed: multiple serialised objects and object references within a TCP stream, ambiguity in mapping interleaved connections to concurrent protocol sessions, and semantic validation of candidate reconstructions. We evaluate the methodology on 121 protocols from the AnBx test suite. For single-session captures, the standard JDK cryptographic provider recovers 97.8% of expected objects and achieves 94.2% successful reconstruction, where success requires complete object recovery and a clean passive-intruder execution. For the 112 parallel-capable protocols, the toolchain achieves 89.3% successful reconstruction for both two and three concurrent sessions, while recovering approximately 97.2% of expected objects. Experiments with mismatched protocol-role mappings and deliberately mutated traces show that the generated Dolev-Yao checks respond to semantic inconsistencies. By making the reconstruction step explicit, the solution produces protocol-level observations on which assessments of protocol resilience and the design of defensive countermeasures can be built.

## Well, That Wasn't Fun: October 16 '26

DevFeed: [Well, That Wasn't Fun: October 16 '26](<https://devfeed.tech/articles/well-that-wasn-t-fun-october-16-26-121085.md>)

Original publisher: [Read original article](<https://fly.io/blog/well-that-wasnt-fun/>)

Author: Daniel Botha

Published: 2026-11-06T00:00:00Z

Content type: news

Language: en

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

Topics: [fly.io](<https://devfeed.tech/topics/fly-io.md>), [virtual machines](<https://devfeed.tech/topics/virtual-machines.md>), [virtualization](<https://devfeed.tech/topics/virtualization.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [payload](<https://devfeed.tech/topics/payload.md>), [network-intelligence](<https://devfeed.tech/topics/network-intelligence.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [based](<https://devfeed.tech/tags/based.md>), [bug](<https://devfeed.tech/tags/bug.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [card](<https://devfeed.tech/tags/card.md>), [cdn](<https://devfeed.tech/tags/cdn.md>), [chaos](<https://devfeed.tech/tags/chaos.md>), [checksums](<https://devfeed.tech/tags/checksums.md>), [close-to-users](<https://devfeed.tech/tags/close-to-users.md>), [code](<https://devfeed.tech/tags/code.md>), [corruption](<https://devfeed.tech/tags/corruption.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [customer](<https://devfeed.tech/tags/customer.md>), [cve](<https://devfeed.tech/tags/cve.md>), [deploy-app-servers](<https://devfeed.tech/tags/deploy-app-servers.md>), [docker](<https://devfeed.tech/tags/docker.md>), [elixir](<https://devfeed.tech/tags/elixir.md>), [example](<https://devfeed.tech/tags/example.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [field](<https://devfeed.tech/tags/field.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [fly](<https://devfeed.tech/tags/fly.md>), [fly-io](<https://devfeed.tech/tags/fly-io.md>), [fun](<https://devfeed.tech/tags/fun.md>), [heroku-alternative](<https://devfeed.tech/tags/heroku-alternative.md>), [heroku-competitor](<https://devfeed.tech/tags/heroku-competitor.md>), [hosting](<https://devfeed.tech/tags/hosting.md>), [i](<https://devfeed.tech/tags/i.md>), [networking](<https://devfeed.tech/tags/networking.md>), [october](<https://devfeed.tech/tags/october.md>), [postgresql-clusters](<https://devfeed.tech/tags/postgresql-clusters.md>), [servers](<https://devfeed.tech/tags/servers.md>), [the-fly-blog](<https://devfeed.tech/tags/the-fly-blog.md>)

### AI overview

Fly.io patched and rebooted its fleet after CVE-2026-99999 exposed a flaw in Linux IPv4 checksum-offload validation for virtual machine traffic. By redirecting checksum writes, an attacker could alter two trusted IPv4 header bytes after routing and firewall checks, enabling a firewall bypass or a 32-byte out-of-bounds read that might expose host or neighboring VM data. Fly.io says it has no evidence customer data was exposed, but advises customers who ran there October 15-17 to rotate credentials sent over unencrypted protocols, including plaintext traffic over 6PN. The 10-hour rollout required staggered host reboots to limit disruption to Managed Postgres clusters; some machines remained down after hardware and reinstall problems, with no machines or customer data lost. Fly.io also revised incident communications so executives own the communication plan. Its practical security guidance includes encrypting internal traffic, using scoped and expiring agent credentials, limiting shared exposure across trust boundaries, treating agent inputs as hostile, and rehearsing credential rotation.

### Source excerpt

October 16, 2026 began like any other day at Fly.io (meaning, we were anticipating only a medium amount of chaos). By that evening, we had patched and rebooted our entire fleet in response to a Linux kernel bug. The full rollout took 10 hours, but the emotional damage persists. CVE-2026-99999 The rest of this won't make much sense without a clear read on the bug itself, so: the Linux kernel CVE team and pwn.ai announced CVE-2026-99999 on the morning of the 15th. We heard about it that night, and started working on mitigations. IP sleight of hand Protocols like TCP and UDP include checksums to detect corruption on the way across the internet. Calculating one means reading the whole message, which adds up at high speeds. Linux avoids that with checksum offload: instead of calculating the checksum immediately, the kernel stores metadata saying where the checksum should read from and write to. If the checksum turns out not to be needed, the kernel never spends the CPU on it. Not needed, for example, when sending to localhost, or when the network card can add its own checksums. Checksum offload works the same way for virtual machines. The virtual network card accepts checksum metadata and passes it to the host. On the host side, the packet is treated like any other offloaded packet, using the metadata the guest provided. The offload code has checks meant to reject bad metadata. For IPv4, they were wrong. To exploit that, an attacker needs to do three things: point the checksum at a different header field, pick a message that produces the checksum they want, and force the host to calculate the checksum in software. The pwn.ai post goes into detail on pulling that off. The end result: the host receives the packet, parses it, decides where to send it, then applies the checksum, in the wrong place. The attacker now controls two bytes of an IPv4 header the host has already trusted. Based on network activity during the rollout, we believe it's unlikely that any of our customer

## What took the Internet offline in Q3 2026

DevFeed: [What took the Internet offline in Q3 2026](<https://devfeed.tech/articles/what-took-the-internet-offline-in-q3-2026-102566.md>)

Original publisher: [Read original article](<https://blog.cloudflare.com/q3-2026-internet-disruption-summary/>)

Author: Lai Yi Ohlsen

Published: 2026-10-27T13:00:00Z

Content type: article

Language: en

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

Topics: [Internet Trends](<https://devfeed.tech/topics/internet-trends.md>), [Internet Traffic](<https://devfeed.tech/topics/internet-traffic.md>), [Resilience](<https://devfeed.tech/topics/resilience.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Cloudflare](<https://devfeed.tech/topics/cloudflare.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [connectivity](<https://devfeed.tech/tags/connectivity.md>), [government](<https://devfeed.tech/tags/government.md>), [internet](<https://devfeed.tech/tags/internet.md>), [internet-shutdown](<https://devfeed.tech/tags/internet-shutdown.md>), [internet-traffic](<https://devfeed.tech/tags/internet-traffic.md>), [internet-trends](<https://devfeed.tech/tags/internet-trends.md>), [natural-disasters](<https://devfeed.tech/tags/natural-disasters.md>), [outage](<https://devfeed.tech/tags/outage.md>), [power](<https://devfeed.tech/tags/power.md>), [radar](<https://devfeed.tech/tags/radar.md>)

### AI overview

Cloudflare Radar recorded 49 Internet disruptions in Q3 2026. Government-directed shutdowns led with 19 incidents, including 18 scheduled exam shutdowns in Iraq; power outages accounted for 14 incidents across 10 countries, and natural disasters for five. Power failures affected connectivity through grids and facilities: a Honduras interconnection failure disrupted at least 11 providers and cut HTTP traffic 63% for about three hours, while outages in Sevastopol reduced traffic by up to 98%. Storms, earthquakes, conflict, and infrastructure damage also caused regional or localized outages. In Japan, the earthquake's connectivity impact appeared mainly in desktop traffic, with mobile traffic largely unaffected. Recovery varied: 14 disruptions returned to typical traffic levels in under an hour and 16 in under three hours. The account is based on Cloudflare's observations of traffic to its customers; Radar data alone could not establish whether Russia's cable damage was related to elections.

### Source excerpt

An examination of global Internet disruptions observed by Cloudflare Radar during the third quarter of 2026. This review analyzes how power failures, natural disasters, and government shutdowns impacted network connectivity and traffic.

## Smart-TCP: An Agentic AI-based Autonomous and Adaptive TCP Protocol

DevFeed: [Smart-TCP: An Agentic AI-based Autonomous and Adaptive TCP Protocol](<https://devfeed.tech/articles/smart-tcp-an-agentic-ai-based-autonomous-and-adaptive-tcp-protocol-101648.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2512.00491>)

Author: Yule Han, Kezhi Wang, Yizhe Zhao, Kun Yang

Published: 2026-10-27T04:00:00Z

Content type: article

Language: en

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

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [arithmetic](<https://devfeed.tech/tags/arithmetic.md>), [atomic](<https://devfeed.tech/tags/atomic.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [classification](<https://devfeed.tech/tags/classification.md>), [completion](<https://devfeed.tech/tags/completion.md>), [context](<https://devfeed.tech/tags/context.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [fallback](<https://devfeed.tech/tags/fallback.md>), [fast](<https://devfeed.tech/tags/fast.md>), [feature](<https://devfeed.tech/tags/feature.md>), [framework](<https://devfeed.tech/tags/framework.md>), [full](<https://devfeed.tech/tags/full.md>), [history](<https://devfeed.tech/tags/history.md>), [tcp](<https://devfeed.tech/tags/tcp.md>)

### AI overview

Smart-TCP separates learned control decisions from deterministic TCP arithmetic: a classifier sends routine segments to a lightweight SLM fast path and anomalous or boundary cases to an LLM slow path; an ALU computes sequence and acknowledgment numbers, session state tracks connection context, and a Protocol Guard checks candidate actions against RFC-derived rules. In reported tests, it achieved 99.14% action accuracy on 700 slow-path anomaly and fallback samples, 100% full-lifecycle success across 300 ideal fast-path session windows, and 96.5% completion for ordinary 32-KiB transfers at 20% segment loss. These results are limited to the evaluated scenarios.

### Source excerpt

arXiv:2512.00491v4 Announce Type: replace Abstract: The Transmission Control Protocol (TCP) relies on deterministic state machines and sequence-number arithmetic to ensure reliable communication. However, hard-coded protocol logic is difficult to adapt to increasingly complex and dynamic network conditions. This paper proposes Smart-TCP, an agentic AI-based transport protocol framework that organizes TCP control logic as a fast/slow model-assisted decision process. Specifically, a feature-aware classifier routes routine segments to a lightweight Small Language Model (SLM) fast path and anomalous or boundary cases to a Large Language Model (LLM) slow path, while an Arithmetic Logic Unit (ALU) handles deterministic computation of sequence and acknowledgment numbers. A state module maintains connection state, packet history, and key control variables, enabling protocol decisions over the full session context. A scoped Protocol Guard checks candidate decisions against RFC-derived rules before execution. We evaluate Smart-TCP through path classification, atomic packet processing, slow-path anomaly response, and end-to-end session interaction. Experimental results show that Smart-TCP achieves 99.14% action accuracy on 700 slow-path anomaly and fallback samples, and a 100% full-lifecycle success rate over 300 ideal fast-path end-to-end session windows. The ordinary 32-KiB transfer tests achieve a 96.5% completion rate at 20% segment loss. These results suggest that separating model-based protocol decisions from deterministic arithmetic improves the reliability of TCP control for network service operation.

## Chasing that 32 ms tail latency in ClickHouse

DevFeed: [Chasing that 32 ms tail latency in ClickHouse](<https://devfeed.tech/articles/chasing-that-32-ms-tail-latency-in-clickhouse-85878.md>)

Original publisher: [Read original article](<https://clickhouse.com/blog/chasing-32ms-tail-latency-clickhouse>)

Author: Kavi Kanagaraj

Published: 2026-10-13T08:49:51Z

Content type: article

Language: en

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

Topics: [scylladb](<https://devfeed.tech/topics/scylladb.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Latency](<https://devfeed.tech/topics/latency.md>)

Tags: [batching](<https://devfeed.tech/tags/batching.md>), [clickhouse](<https://devfeed.tech/tags/clickhouse.md>), [latency](<https://devfeed.tech/tags/latency.md>), [network](<https://devfeed.tech/tags/network.md>), [tcp](<https://devfeed.tech/tags/tcp.md>)

### AI overview

The article traces an extra ~32 ms of tail latency on simple ClickHouse queries after idle periods to metadata packets exhausting a reset TCP congestion window. It describes reproducing the issue with simulated network delay and reducing the extra latency by batching metadata packets at the application layer.

### Source excerpt

A customer saw an extra ~32 ms on simple SELECT queries after idle periods. We traced it to unbatched Native protocol metadata packets overflowing a reset TCP congestion window, reproduced it with network namespaces, and fixed it by batching them.

## Original RFC1149 Packet Up For Auction At Christies

DevFeed: [Original RFC1149 Packet Up For Auction At Christies](<https://devfeed.tech/articles/original-rfc1149-packet-up-for-auction-at-christies-85037.md>)

Original publisher: [Read original article](<https://entertainment.slashdot.org/story/26/10/11/0050222/original-rfc1149-packet-up-for-auction-at-christies>)

Author: EditorDavid

Published: 2026-10-11T15:34:00Z

Content type: news

Language: en

Sources: [Slashdot](<https://devfeed.tech/sources/slashdot.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [2600](<https://devfeed.tech/tags/2600.md>), [auction](<https://devfeed.tech/tags/auction.md>), [badger](<https://devfeed.tech/tags/badger.md>), [carrier](<https://devfeed.tech/tags/carrier.md>), [here](<https://devfeed.tech/tags/here.md>), [humor](<https://devfeed.tech/tags/humor.md>), [internet](<https://devfeed.tech/tags/internet.md>), [ip](<https://devfeed.tech/tags/ip.md>), [original](<https://devfeed.tech/tags/original.md>), [over](<https://devfeed.tech/tags/over.md>), [packet](<https://devfeed.tech/tags/packet.md>), [packets](<https://devfeed.tech/tags/packets.md>), [story](<https://devfeed.tech/tags/story.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [tcp-ip](<https://devfeed.tech/tags/tcp-ip.md>), [time](<https://devfeed.tech/tags/time.md>), [unique](<https://devfeed.tech/tags/unique.md>), [up](<https://devfeed.tech/tags/up.md>)

### AI overview

A preserved packet from a 2001 implementation of RFC 1149, which carried TCP/IP over carrier pigeon, is being auctioned at Christie's. The reported current bid is USD 2,600.

### Source excerpt

Long-time Slashdot reader badger.foo writes: As some of you may remember, in 2001 RFC1149 (TCP/IP over carrier pigeon) was implemented. Now one of the packets preserved from that event is up for auction at Christies. Direct link to the auction (current bid: USD 2600) is here. This is a unique chance to own a piece of Internet history! Read more of this story at Slashdot.

## What Is an H.264 Streaming Encoder? Settings, Protocols, and Best Practices

DevFeed: [What Is an H.264 Streaming Encoder? Settings, Protocols, and Best Practices](<https://devfeed.tech/articles/what-is-an-h-264-streaming-encoder-settings-protocols-and-best-practices-87572.md>)

Original publisher: [Read original article](<https://www.dacast.com/blog/what-is-h-264-streaming-encoder/>)

Author: Jon Whitehead

Published: 2026-10-05T00:35:00Z

Content type: tutorial

Language: en

Sources: [DaCast](<https://devfeed.tech/sources/dacast.md>)

Topics: [Web](<https://devfeed.tech/topics/web.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [best-practices](<https://devfeed.tech/tags/best-practices.md>), [h-264](<https://devfeed.tech/tags/h-264.md>), [protocols](<https://devfeed.tech/tags/protocols.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [the-video-experts-blog](<https://devfeed.tech/tags/the-video-experts-blog.md>)

### AI overview

A practical guide to H.264 video encoding for live and on-demand streaming. It explains encoding workflows, bitrate and encoder choices, protocol compatibility, and how H.264 compares with newer codecs, emphasizing its broad device support and balance of quality and bandwidth use.

### Source excerpt

By Dacast Editorial Team | Reviewed by Jon Whitehead, COO at Dacast | Updated October 2026. If you stream video online, chances are you're using an H.264 streaming encoder, whether you know it or not. In 2026, H.264 remains one of the most widely supported video codecs for live and on-demand streaming. It delivers a solid [...] The post What Is an H.264 Streaming Encoder? Settings, Protocols, and Best Practices appeared first on Dacast.

## Hedge 321: Sockets

DevFeed: [Hedge 321: Sockets](<https://devfeed.tech/articles/hedge-321-sockets-64005.md>)

Original publisher: [Read original article](<https://rule11.tech/hedge-321/>)

Author: Russ

Published: 2026-10-02T12:50:25Z

Content type: article

Language: en

Sources: [rule 11 reader](<https://devfeed.tech/sources/rule-11-reader.md>)

Topics: [Internet](<https://devfeed.tech/topics/internet.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [audio](<https://devfeed.tech/tags/audio.md>), [dns](<https://devfeed.tech/tags/dns.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [hedge](<https://devfeed.tech/tags/hedge.md>), [interface](<https://devfeed.tech/tags/interface.md>), [internet](<https://devfeed.tech/tags/internet.md>), [sockets](<https://devfeed.tech/tags/sockets.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [tcp-ip](<https://devfeed.tech/tags/tcp-ip.md>)

### AI overview

This podcast discusses the history and importance of the socket interface in internet technologies.

### Source excerpt

There are many "unsung technologies" in the Internet. While we often think of TCP/IP, various models, Ethernet, and DNS, we don't often think about the lowly socket interface. What is the history of the socket, and why is it important in the world of the Internet? George Michaelson joins Colin Doyle and Russ White to discuss the history and importance of sockets. https://media.blubrry.com/hedge/media.blubrry.com/hedge/content.blubrry.com/hedge/hedge-321.mp3

## Redirecting a PS5's RTMP broadcast stream to a local receiver

DevFeed: [Redirecting a PS5's RTMP broadcast stream to a local receiver](<https://devfeed.tech/articles/hijacking-the-ps5-s-rtmp-stream-61878.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/09/29/hijacking-the-ps5s-rtmp-stream/>)

Author: Anne Barela

Published: 2026-09-29T16:03:18Z

Content type: article

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Domain Name System (DNS)](<https://devfeed.tech/topics/dns-resolution-process.md>), [DNS security](<https://devfeed.tech/topics/dns-security.md>), [vp9](<https://devfeed.tech/topics/vp9.md>), [Network Attached Storage (NAS)](<https://devfeed.tech/topics/network-attached-storage-nas.md>)

Tags: [arduino](<https://devfeed.tech/tags/arduino.md>), [discord](<https://devfeed.tech/tags/discord.md>), [dns](<https://devfeed.tech/tags/dns.md>), [gameplay](<https://devfeed.tech/tags/gameplay.md>), [gaming](<https://devfeed.tech/tags/gaming.md>), [h-264](<https://devfeed.tech/tags/h-264.md>), [hijacking](<https://devfeed.tech/tags/hijacking.md>), [nginx](<https://devfeed.tech/tags/nginx.md>), [openwrt](<https://devfeed.tech/tags/openwrt.md>), [openwrt-router](<https://devfeed.tech/tags/openwrt-router.md>), [ps5](<https://devfeed.tech/tags/ps5.md>), [raspberry-pi](<https://devfeed.tech/tags/raspberry-pi.md>), [streaming](<https://devfeed.tech/tags/streaming.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

The article describes routing a PS5 broadcast stream to a local receiver by spoofing DNS for a plain RTMP ingest domain. Using dnsmasq, nginx-rtmp, and an OpenWRT router, the setup receives 1080p60 H.264 video with AAC stereo for playback with less than a second of delay.

### Source excerpt

Yash Garg wanted to share PS5 gameplay to Discord without buying a capture card, and wrote up the result on yashgarg.dev. The PS5 has a Broadcast button that pushes video to Twitch or YouTube over RTMP, and it finds those servers by DNS lookup. Control the DNS and the stream goes wherever one wants. The [...]

## A compilation of W3C in Japan's 30th anniversary events

DevFeed: [A compilation of W3C in Japan's 30th anniversary events](<https://devfeed.tech/articles/a-compilation-of-w3c-in-japan-s-30th-anniversary-events-49171.md>)

Original publisher: [Read original article](<https://www.w3.org/blog/2026/a-compilation-of-w3c-in-japans-30th-anniversary-events/>)

Author: Naomi Yoshizawa

Published: 2026-09-09T20:50:00Z

Content type: article

Language: en

Sources: [W3C](<https://devfeed.tech/sources/w3c-blog.md>)

Topics: [W3C](<https://devfeed.tech/topics/w3c.md>), [Web](<https://devfeed.tech/topics/web.md>), [browser](<https://devfeed.tech/topics/browser.md>), [coding-community](<https://devfeed.tech/topics/coding-community.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [protocols](<https://devfeed.tech/topics/protocols.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [anniversary](<https://devfeed.tech/tags/anniversary.md>), [blogs](<https://devfeed.tech/tags/blogs.md>), [browser](<https://devfeed.tech/tags/browser.md>), [community](<https://devfeed.tech/tags/community.md>), [corporate](<https://devfeed.tech/tags/corporate.md>), [developer-community](<https://devfeed.tech/tags/developer-community.md>), [events](<https://devfeed.tech/tags/events.md>), [japan](<https://devfeed.tech/tags/japan.md>), [protocols](<https://devfeed.tech/tags/protocols.md>), [tcp-ip](<https://devfeed.tech/tags/tcp-ip.md>), [w3c](<https://devfeed.tech/tags/w3c.md>), [web](<https://devfeed.tech/tags/web.md>)

### AI overview

W3C Japan's 30th anniversary activities included a member event, a public exhibition, and the "Web Together" meetup for W3C members and Japan's web developer community. The meetup featured lightning talks and a panel reflecting on the web's history, value, challenges, and future.

### Source excerpt

2026 marks the 30th anniversary of the establishment of W3C in Japan. As part of these commemorative activities, W3C Japan held an event for W3C members in May, an exhibition for the general public in June and an event early September for W3C members and the web developer community in Japan.

## IP Addressing, Part III: Inside the IP Packet

DevFeed: [IP Addressing, Part III: Inside the IP Packet](<https://devfeed.tech/articles/ip-addressing-part-iii-inside-the-ip-packet-18196.md>)

Original publisher: [Read original article](<https://newsletter.francofernando.com/p/ip-addressing-part-iii-inside-the>)

Author: Franco Fernando

Published: 2026-08-07T09:49:52Z

Content type: article

Language: en

Sources: [The Polymathic Engineer](<https://devfeed.tech/sources/the-polymathic-engineer.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [payload](<https://devfeed.tech/topics/payload.md>), [MTU mismatch](<https://devfeed.tech/topics/mtu-mismatch.md>)

Tags: [fragmentation](<https://devfeed.tech/tags/fragmentation.md>), [header](<https://devfeed.tech/tags/header.md>), [ip](<https://devfeed.tech/tags/ip.md>), [ipv4](<https://devfeed.tech/tags/ipv4.md>), [ipv6](<https://devfeed.tech/tags/ipv6.md>), [mtu](<https://devfeed.tech/tags/mtu.md>), [packet](<https://devfeed.tech/tags/packet.md>)

### AI overview

The excerpt introduces IPv4 datagrams as a header, usually 20 bytes, followed by a payload, then explains that the header tells routers how to interpret the packet and receivers where the payload begins. It identifies the version field (4 for IPv4), the variable header length due to optional fields, and the 16-bit total-length field, which allows up to 65,535 bytes in theory. The text says practical datagrams are rarely larger than 1,500 bytes and previews fragmentation, MTU, checksums at both TCP and IP layers, and IPv6 redesign, but the excerpt ends before explaining those topics.

### Source excerpt

What actually travels on the wire: the IPv4 header, fragmentation, and what IPv6 changed.

## Putting the Tech in Fintech: How We Use FIX to Execute Trades at Scale

DevFeed: [Putting the Tech in Fintech: How We Use FIX to Execute Trades at Scale](<https://devfeed.tech/articles/putting-the-tech-in-fintech-how-we-use-fix-to-execute-trades-at-scale-20463.md>)

Original publisher: [Read original article](<https://eng.wealthfront.com/2026/04/21/putting-the-tech-in-fintech-how-we-use-fix-to-execute-trades-at-scale/>)

Author: Grace Sun

Published: 2026-04-22T00:27:03Z

Content type: article

Language: en

Sources: [Wealthfront](<https://devfeed.tech/sources/wealthfront.md>)

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [Network](<https://devfeed.tech/topics/network.md>), [workflow automation](<https://devfeed.tech/topics/workflow-automation.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [2023](<https://devfeed.tech/tags/2023.md>), [application](<https://devfeed.tech/tags/application.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automation](<https://devfeed.tech/tags/automation.md>), [backend](<https://devfeed.tech/tags/backend.md>), [clients](<https://devfeed.tech/tags/clients.md>), [complex](<https://devfeed.tech/tags/complex.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [engine](<https://devfeed.tech/tags/engine.md>), [financial-technology](<https://devfeed.tech/tags/financial-technology.md>), [fintech](<https://devfeed.tech/tags/fintech.md>), [frontend](<https://devfeed.tech/tags/frontend.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [scale](<https://devfeed.tech/tags/scale.md>), [systems](<https://devfeed.tech/tags/systems.md>), [tech](<https://devfeed.tech/tags/tech.md>), [technology](<https://devfeed.tech/tags/technology.md>), [trades](<https://devfeed.tech/tags/trades.md>), [trading](<https://devfeed.tech/tags/trading.md>), [us](<https://devfeed.tech/tags/us.md>), [volumes](<https://devfeed.tech/tags/volumes.md>), [wealthfront-engineering](<https://devfeed.tech/tags/wealthfront-engineering.md>)

### AI overview

Wealthfront describes trade execution at scale as a layered system that automates broker communication, responding to the volume and speed of modern US equity trading. It compares the architecture to a phone call: networking provides the connection, FIX the message language, QuickFIX/J manages the session, and an internal application layer interprets and translates messages. The excerpt details the networking layer: Wealthfront maintains a dedicated connection to each broker, uses standard IP routing to provisioned broker addresses, and introduces the TCP/IP stack as its model. The excerpt ends mid-description, so it does not establish the remaining network details or explain how FIX, QuickFIX/J, and the application layer work in practice.

### Source excerpt

At Wealthfront, the "tech" in financial technology isn't just a buzzword--it's the foundation of everything we build. Beneath the intuitive frontend our clients interact with lies a complex ecosystem of distributed systems. One of the most critical pieces of that backend architecture is the engine that enables us to route massive volumes of trades efficiently. ... Read more

## Pre-Authentication Buffer Overflow in GNU inetutils Telnetd Enables Remote Code Execution

DevFeed: [Pre-Authentication Buffer Overflow in GNU inetutils Telnetd Enables Remote Code Execution](<https://devfeed.tech/articles/a-32-year-old-bug-walks-into-a-telnet-server-gnu-inetutils-telnetd-cve-2026-32746-pre-auth-rce-50780.md>)

Original publisher: [Read original article](<https://labs.watchtowr.com/a-32-year-old-bug-walks-into-a-telnet-server-gnu-inetutils-telnetd-cve-2026-32746/>)

Author: McCaulay Hudson

Published: 2026-03-19T20:21:07Z

Content type: article

Language: en

Sources: [watchTowr Labs](<https://devfeed.tech/sources/watchtowr-labs.md>)

Topics: [Vulnerabilities](<https://devfeed.tech/topics/vulnerabilities.md>), [Exploit](<https://devfeed.tech/topics/exploit.md>), [Security research](<https://devfeed.tech/topics/security-research.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [Protocol (disambiguation)](<https://devfeed.tech/topics/protocol.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Network](<https://devfeed.tech/topics/network.md>)

Tags: [cve](<https://devfeed.tech/tags/cve.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [linux](<https://devfeed.tech/tags/linux.md>), [network-protocol](<https://devfeed.tech/tags/network-protocol.md>), [rce](<https://devfeed.tech/tags/rce.md>), [security](<https://devfeed.tech/tags/security.md>), [security-research](<https://devfeed.tech/tags/security-research.md>), [tcp-ip](<https://devfeed.tech/tags/tcp-ip.md>), [unix](<https://devfeed.tech/tags/unix.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

The article examines CVE-2026-32746, a pre-authentication BSS-based buffer overflow in the LINEMODE SLC negotiation handler of GNU inetutils Telnetd. It discusses Telnet's insecure legacy design, the vulnerability's potential impact, affected implementations, and the challenges of exploitation.

### Source excerpt

A long, long time ago, in a land free of binary exploit mitigations, when Unix still roamed the Earth, there lived a pre-authentication Telnetd vulnerability. In fact, this vulnerability was born so long ago (way back in 1994) that it may even be older than you. To put the

## Won't You Be My Neighbor? Part 2: The Multi-Protocol Agent

DevFeed: [Won't You Be My Neighbor? Part 2: The Multi-Protocol Agent](<https://devfeed.tech/articles/won-t-you-be-my-neighbor-part-2-the-multi-protocol-agent-60923.md>)

Original publisher: [Read original article](<https://automateyournetwork.com/uncategorized/wont-you-be-my-neighbor-part-2-the-multi-protocol-agent/>)

Author: John Capobianco

Published: 2026-01-18T12:00:00Z

Content type: tutorial

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [IP routing](<https://devfeed.tech/topics/ip-routing.md>), [BGP](<https://devfeed.tech/topics/bgp.md>), [Routing Security](<https://devfeed.tech/topics/routing-security.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [cursor](<https://devfeed.tech/topics/cursor.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [article](<https://devfeed.tech/tags/article.md>), [bgp](<https://devfeed.tech/tags/bgp.md>), [ipv6](<https://devfeed.tech/tags/ipv6.md>), [multi-protocol](<https://devfeed.tech/tags/multi-protocol.md>), [ospf](<https://devfeed.tech/tags/ospf.md>), [rpki](<https://devfeed.tech/tags/rpki.md>)

### AI overview

The article describes extending an AI network agent beyond OSPF to support multiple routing protocols, including iBGP, eBGP, MP-BGP for IPv6, RPKI validation, FlowSpec, and OSPFv3. It says these capabilities operate in one agent with adjacencies to commercial routers, then begins discussing BGP route reflection and session establishment.

### Source excerpt

## Or: How I Taught an AI Agent to Speak Every Language in the Network *Building on [Part 1: Teaching an AI to Speak OSPF](https://www.automateyournetwork.ca/uncategorized/i-taught-an-ai-agent-to-speak-ospf-its-now-my-routers-neighbour/)* --

## Learn How to Lower Heroku Dyno Latency through Persistent Connections (Keep-alive)

DevFeed: [Learn How to Lower Heroku Dyno Latency through Persistent Connections (Keep-alive)](<https://devfeed.tech/articles/learn-how-to-lower-heroku-dyno-latency-through-persistent-connections-keep-alive-26462.md>)

Original publisher: [Read original article](<https://www.heroku.com/blog/learn-how-to-lower-heroku-dyno-latency-keep-alive/>)

Author: Richard Schneeman

Published: 2025-11-04T16:27:37Z

Content type: article

Language: en

Sources: [Heroku](<https://devfeed.tech/sources/heroku.md>)

Topics: [Latency](<https://devfeed.tech/topics/latency.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [Heroku](<https://devfeed.tech/topics/heroku.md>), [Web app](<https://devfeed.tech/topics/webapp.md>), [Network](<https://devfeed.tech/topics/network.md>)

Tags: [engineering](<https://devfeed.tech/tags/engineering.md>), [heroku](<https://devfeed.tech/tags/heroku.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [http](<https://devfeed.tech/tags/http.md>), [keep](<https://devfeed.tech/tags/keep.md>), [latency](<https://devfeed.tech/tags/latency.md>), [performance](<https://devfeed.tech/tags/performance.md>), [performance-optimization](<https://devfeed.tech/tags/performance-optimization.md>), [persistence](<https://devfeed.tech/tags/persistence.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [transmission](<https://devfeed.tech/tags/transmission.md>), [web](<https://devfeed.tech/tags/web.md>), [web-applications](<https://devfeed.tech/tags/web-applications.md>)

### AI overview

Keep-alive connections can reduce request latency by reusing an established TCP connection for later requests, amortizing connection setup and avoiding the TCP slow-start phase for data sent over that connection. The excerpt contrasts this with HTTP/1.0's one-connection-per-request pattern, which repeatedly opened and closed connections. It frames repeated TCP slow starts as a performance bottleneck between Heroku's router and application dynos; the excerpt does not include implementation steps or details of the router improvements.

### Source excerpt

The Performance Penalty of Repeated Connections Before the latest improvements to the Heroku Router, every connection between the router and your application dyno risked incurring the latency penalty of a TCP slow start. To understand why this is a performance bottleneck for modern web applications, we must look at the fundamentals of the Transmission Control [...] The post Learn How to Lower Heroku Dyno Latency through Persistent Connections (Keep-alive) appeared first on Heroku.

## What we've been reading in March (2025)

DevFeed: [What we've been reading in March (2025)](<https://devfeed.tech/articles/what-we-ve-been-reading-in-march-2025-47304.md>)

Original publisher: [Read original article](<https://interrupt.memfault.com/blog/march-2025-roundup>)

Author: Gillian Minnehan

Published: 2025-04-03T00:00:00Z

Content type: article

Language: en

Sources: [Interrupt](<https://devfeed.tech/sources/interrupt.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [ESP32](<https://devfeed.tech/topics/esp32.md>), [Wi-Fi](<https://devfeed.tech/topics/wi-fi.md>), [Bluetooth](<https://devfeed.tech/topics/bluetooth.md>), [Linux](<https://devfeed.tech/topics/linux.md>), [Rust](<https://devfeed.tech/topics/rust.md>), [Memory Safety](<https://devfeed.tech/topics/memory-safety.md>), [Security](<https://devfeed.tech/topics/security.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Ethernet](<https://devfeed.tech/topics/ethernet.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [bluetooth](<https://devfeed.tech/tags/bluetooth.md>), [build](<https://devfeed.tech/tags/build.md>), [embedded](<https://devfeed.tech/tags/embedded.md>), [esp32](<https://devfeed.tech/tags/esp32.md>), [ethernet](<https://devfeed.tech/tags/ethernet.md>), [for-devs](<https://devfeed.tech/tags/for-devs.md>), [hacking](<https://devfeed.tech/tags/hacking.md>), [linux](<https://devfeed.tech/tags/linux.md>), [memory-safety](<https://devfeed.tech/tags/memory-safety.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [programming](<https://devfeed.tech/tags/programming.md>), [reading](<https://devfeed.tech/tags/reading.md>), [rust](<https://devfeed.tech/tags/rust.md>)

### AI overview

A March 2025 roundup of articles, videos, and tools covering embedded networking, ESP32 Wi-Fi and Bluetooth, embedded Linux and SBOMs, Rust, memory safety, hardware security, and KiCad.

### Source excerpt

Here are the articles, videos, and tools that we've been excited about this March. Continue reading...

## Comparing IP and CLNP: Reaching Off-Subnet Nodes

DevFeed: [Comparing IP and CLNP: Reaching Off-Subnet Nodes](<https://devfeed.tech/articles/comparing-ip-and-clnp-reaching-off-subnet-nodes-11149.md>)

Original publisher: [Read original article](<https://blog.ipspace.net/2025/03/comparing-ip-clnp-off-subnet-nodes/>)

Published: 2025-03-05T07:08:00Z

Content type: tutorial

Language: en

Sources: [ipSpace](<https://devfeed.tech/sources/ipspace-net-blog-2.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Network](<https://devfeed.tech/topics/network.md>), [DHCP](<https://devfeed.tech/topics/dhcp.md>), [Load Balancing](<https://devfeed.tech/topics/load-balancing.md>)

Tags: [cisco](<https://devfeed.tech/tags/cisco.md>), [dhcp](<https://devfeed.tech/tags/dhcp.md>), [ip](<https://devfeed.tech/tags/ip.md>), [ipv4](<https://devfeed.tech/tags/ipv4.md>), [ipv6](<https://devfeed.tech/tags/ipv6.md>), [load-balancing](<https://devfeed.tech/tags/load-balancing.md>), [networking-fundamentals](<https://devfeed.tech/tags/networking-fundamentals.md>), [routing](<https://devfeed.tech/tags/routing.md>)

### AI overview

This article compares how TCP/IP and CLNP identify a first-hop router for traffic to off-subnet nodes. It describes IPv4 default routes, DHCP configuration, first-hop redundancy protocols, and CLNP Intermediate System Hellos.

### Source excerpt

The previous blog post in this series discussed how TCP/IP and CLNP reach adjacent nodes and build ARP/ND/ES caches. Now let's move one step further: how do nodes running IPv4/IPv6 or CLNP discover the first-hop router that could forward their traffic to off-subnet nodes they want to communicate with? Read more ...

## It's always TCP\_NODELAY. Every damn time.

DevFeed: [It's always TCP\_NODELAY. Every damn time.](<https://devfeed.tech/articles/it-s-always-tcp-nodelay-every-damn-time-12557.md>)

Original publisher: [Read original article](<http://brooker.co.za/blog/2024/05/09/nagle.html>)

Author: Marc Brooker

Published: 2024-05-09T00:00:00Z

Content type: article

Language: en

Sources: [Marc Brooker's](<https://devfeed.tech/sources/marc-brooker-s-blog.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Network](<https://devfeed.tech/topics/network.md>), [Service fabric](<https://devfeed.tech/topics/service-fabric.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [delayed-ack](<https://devfeed.tech/tags/delayed-ack.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [headers](<https://devfeed.tech/tags/headers.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [latency](<https://devfeed.tech/tags/latency.md>), [nagle](<https://devfeed.tech/tags/nagle.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [packets](<https://devfeed.tech/tags/packets.md>), [socket](<https://devfeed.tech/tags/socket.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [throughput](<https://devfeed.tech/tags/throughput.md>), [timer](<https://devfeed.tech/tags/timer.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

The excerpt recommends enabling TCP_NODELAY for latency-sensitive distributed systems: Nagle's algorithm can hold outgoing data until earlier data is acknowledged, and delayed ACKs can compound the wait. The author argues that modern distributed applications usually send larger messages, with tiny-message efficiency better handled at the application layer, so waiting even one round-trip may not be worthwhile. Disabling Nagle may make code that writes one byte at a time less efficient; the author says such code should be fixed. The excerpt also notes TCP_QUICKACK as a less attractive alternative because of portability and unusual semantics. The source is truncated, so it does not establish the full discussion.

### Source excerpt

It's always TCP_NODELAY. Every damn time. It's not the 1980s anymore, thankfully. The first thing I check when debugging latency issues in distributed systems is whether TCP_NODELAY is enabled. And it's not just me. Every distributed system builder I know has lost hours to latency issues quickly fixed by enabling this simple socket option, suggesting that the default behavior is wrong, and perhaps that the whole concept is outmoded. First, let's be clear about what we're talking about. There's no better source than John Nagle's RFC896 from 19841. First, the problem statement: There is a special problem associated with small packets. When TCP is used for the transmission of single-character messages originating at a keyboard, the typical result is that 41 byte packets (one byte of data, 40 bytes of header) are transmitted for each byte of useful data. This 4000% overhead is annoying but tolerable on lightly loaded networks. In short, Nagle was interested in better amortizing the cost of TCP headers, to get better throughput out of the network. Up to 40x better throughput! These tiny packets had two main causes: human-interactive applications like shells, where folks were typing a byte at a time, and poorly implemented programs that dribbled messages out to the kernel through many write calls. Nagle's proposal for fixing this was simple and smart: A simple and elegant solution has been discovered. The solution is to inhibit the sending of new TCP segments when new outgoing data arrives from the user if any previously transmitted data on the connection remains unacknowledged. When many people talk about Nagle's algorithm, they talk about timers, but RFC896 doesn't use any kind of timer other than the round-trip time on the network. Nagle's Algorithm and Delayed Acks Nagle's nice, clean, proposal interacted poorly with another TCP feature: delayed ACK. The idea behind delayed ACK is to delay sending the acknowledgement of a packet at least until there's some data to sen

## Understanding TCP The Three Way Handshake - Make completely with AI

DevFeed: [Understanding TCP The Three Way Handshake - Make completely with AI](<https://devfeed.tech/articles/understanding-tcp-the-three-way-handshake-make-completely-with-ai-66362.md>)

Original publisher: [Read original article](<https://www.youtube.com/watch?v=zk5cwVAy9nI>)

Published: 2024-01-19T12:00:00Z

Content type: tutorial

Language: en

Sources: [Automate Your Network](<https://devfeed.tech/sources/automate-your-network.md>)

Topics: [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [TLS handshake](<https://devfeed.tech/topics/tls-handshake.md>)

Tags: [camera-phone](<https://devfeed.tech/tags/camera-phone.md>), [free](<https://devfeed.tech/tags/free.md>), [handshake](<https://devfeed.tech/tags/handshake.md>), [sharing](<https://devfeed.tech/tags/sharing.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [upload](<https://devfeed.tech/tags/upload.md>), [video](<https://devfeed.tech/tags/video.md>), [video-phone](<https://devfeed.tech/tags/video-phone.md>)

### AI overview

A video about the TCP three-way handshake. Its description says it was made with InVideo AI from a single prompt.

### Source excerpt

John Capobianco - Video publication

## The origins of the Idle Scan

DevFeed: [The origins of the Idle Scan](<https://devfeed.tech/articles/the-origins-of-the-idle-scan-20630.md>)

Original publisher: [Read original article](<http://antirez.com/news/139>)

Author: Antirez

Published: 2023-10-19T10:40:27Z

Content type: article

Language: en

Sources: [Antirez](<https://devfeed.tech/sources/antirez.md>)

Topics: [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [attack](<https://devfeed.tech/tags/attack.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [counter](<https://devfeed.tech/tags/counter.md>), [field](<https://devfeed.tech/tags/field.md>), [network](<https://devfeed.tech/tags/network.md>), [operating-systems](<https://devfeed.tech/tags/operating-systems.md>), [packet](<https://devfeed.tech/tags/packet.md>), [packets](<https://devfeed.tech/tags/packets.md>), [practical](<https://devfeed.tech/tags/practical.md>), [research](<https://devfeed.tech/tags/research.md>), [scan](<https://devfeed.tech/tags/scan.md>), [tcp](<https://devfeed.tech/tags/tcp.md>), [tcp-ip](<https://devfeed.tech/tags/tcp-ip.md>), [theoretical](<https://devfeed.tech/tags/theoretical.md>), [tool](<https://devfeed.tech/tags/tool.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [udp](<https://devfeed.tech/tags/udp.md>), [well-known](<https://devfeed.tech/tags/well-known.md>)

### AI overview

The author traces Idle scan to a 1998 observation while experimenting with Hping: some operating systems incremented a shared IP ID counter for every outgoing packet. That behavior could reveal a computer's outgoing traffic, a risk others reportedly considered negligible, but the author suspected it could enable a more serious attack. After reasoning through the method with Lorenzo Cavallaro, they modified Hping, tested the theory successfully, and announced it on BUGTRAQ. The account recalls that the first announcement proposed "dumb host scan"; the name Idle scan later prevailed.

### Source excerpt

The Idle scan was conceived at the end of 1998, evidenced by emails. I had moved to Milan a few months prior, having been there since September if I recall correctly, brimming with new ideas, unaware that my stay in that city would be brief. I spent the summer on the beaches of Sicily, mainly occupied with reading many books recommended by the folks at Seclab (mostly by David). However, those readings needed a catalyst: the Idle scan was an attack born from theoretical rumination, but the stream of thoughts originated from a rather practical circumstance. I had recently created Hping, a tool whose logo was borrowed from that of Nutella. I mention this to emphasize the seriousness that governed my efforts at that time -- after all, I was only twenty-one and already in Northern Italy with a full-time job on my shoulders; some understanding was warranted. Hping was a Swiss Army knife for the TCP/IP protocol. Its initial use was mostly exploratory, for research. With Hping, you could assemble TCP, UDP, and ICMP packets in the most bizarre manner, and encapsulate them in equally eccentric IP packets, fragmented, with fields set to anomalous values. These packets were sent around to observe the network stack response of different operating systems. This is where Idle scan originates: playing with Hping for just a few minutes revealed a well-known yet (to me) surprising fact. The response packets had an ID field that continuously incremented by some measure. At that time, given that the attacks I would later disclose were not yet known, this ID field behavior aroused no concern. Every time an operating system emitted an IP packet, it first incremented a counter (which reset to zero once it reached the maximum value of two to the sixteenth power minus one), then the packet was sent with the ID set to the counter's value. The counter was universal for all outgoing packets. This allowed, for starters, to estimate the outgoing traffic of any networked computer. This information

## Why SMTP still matters today

DevFeed: [Why SMTP still matters today](<https://devfeed.tech/articles/why-smtp-still-matters-today-82831.md>)

Original publisher: [Read original article](<https://resend.com/blog/why-smtp-still-matters-today>)

Author: Jonni Lundy

Published: 2023-08-08T00:00:00Z

Content type: release

Language: en

Sources: [Resend](<https://devfeed.tech/sources/resend.md>)

Topics: [email](<https://devfeed.tech/topics/email.md>), [TCP/IP](<https://devfeed.tech/topics/tcp-ip.md>)

Tags: [email](<https://devfeed.tech/tags/email.md>), [guides](<https://devfeed.tech/tags/guides.md>), [smtp](<https://devfeed.tech/tags/smtp.md>)

### AI overview

Resend explains why SMTP remains useful alongside its REST API, citing provider-agnostic integrations and continued use by legacy systems, frameworks, and modern products. The company announces SMTP support and provides setup guidance.

### Source excerpt

The importance of a 40-year-old protocol to the future of email.

[Next page](<https://devfeed.tech/topics/tcp-ip.md?cursor=WyIyMDIzLTA4LTA4VDAwOjAwOjAwKzAwOjAwIiwgIjkxNGZhZDAwLTBjYWQtNGYwYy1hY2ExLTRkMWZjNjc2NjU4NyJd>)