autonomous-systems

4 posts

cloudflare

BGP ORIGIN attribute manipulation and its impact on the Internet (opens in new tab)

BGP’s ORIGIN attribute is intended to describe how a route entered the protocol and should remain unchanged after being set by the originating AS. However, the investigation found that roughly 70% of observed paths had an ORIGIN value different from the original, often because transit providers manipulate it to influence route selection and attract traffic. This widespread practice turns a supposedly stable routing signal into a tool in a revenue-driven competition between networks. ## What the BGP ORIGIN Attribute Means - ORIGIN describes how a route was injected into BGP, rather than identifying the originating Autonomous System. - It has three values: - **IGP (0):** The route originated within the AS. - **EGP (1):** A historical value associated with the obsolete Exterior Gateway Protocol. - **INCOMPLETE (2):** The route was learned through an unknown or external mechanism. - Across routes visible through RIPE RIS and RouteViews: - 89.8% used IGP. - 3.5% used EGP. - 6.7% used INCOMPLETE. - When Local Preference and AS_PATH length are equal, BGP prefers the lower ORIGIN value, making IGP preferable to EGP and INCOMPLETE. - RFC 4271 states that ORIGIN is generated by the originating speaker and **should not be changed** by other speakers. ## How Providers Use ORIGIN Manipulation - A transit provider can rewrite a route’s ORIGIN to IGP, making its path more attractive during BGP route selection. - In the example: - AS64501 announces a route with ORIGIN INCOMPLETE. - AS64502 and AS64503 propagate the route to AS64504. - AS64503 changes the ORIGIN to IGP. - Because both paths have equal AS_PATH lengths, AS64504 selects the path through AS64503. - This manipulation redirects traffic—and potentially transit revenue—toward the provider that altered the attribute. - Some operators also rewrite routes to EGP or INCOMPLETE to make them less preferred than customer routes. - The practice has been discussed at RIPE and LACNIC meetings and has contributed to proposals recommending that ORIGIN be deprecated. ## Measuring ORIGIN Rewriting - The researchers announced: - Three IPv4 prefixes. - Three IPv6 prefixes. - Each prefix used a different ORIGIN value: IGP, EGP, or INCOMPLETE. - Announcements were made from multiple peering locations using BGP Anycast. - After propagation, the prefixes were withdrawn to trigger BGP path hunting, exposing additional routes. - Researchers analyzed: - MRT UPDATE messages from RIPE RIS and RouteViews. - Local BMP data from their border routers. - BGPKIT tools for parsing the data. - UPDATE messages were chosen over routing-table snapshots because they reveal more paths during both announcements and withdrawals. ## Visibility Challenges - Public collectors cannot observe every AS involved in a route’s propagation. - The growth of hyperscalers, CDNs, and direct local peering has flattened the traditional transit hierarchy. - As a result, many paths bypass publicly visible transit networks. - Conclusions about which AS changed an attribute therefore carry some uncertainty, especially beyond directly observed peers. ## Direct-Peer Findings - The first analysis examined two-AS paths such as `ASX AS13335`, where ASX was a direct peer of the researchers. - Because the researchers controlled the original ORIGIN value, any different value observed from a direct peer indicated that peer had rewritten it. - Among 352 IPv4 direct peers: - Three ASes consistently changed routes to EGP. - Four changed routes to INCOMPLETE, potentially attempting to deprioritize them. - One contacted operator confirmed that it rewrote peer- and provider-learned routes to EGP so customer routes would be preferred. - Several ASes advertised both the original ORIGIN and IGP for non-IGP prefixes, likely because they received the routes at multiple locations and altered the value to steer traffic through preferred sites. - Combining these behaviors, the researchers found that nearly 10% of direct peers were rewriting ORIGIN, while the broader experiment detected changes on approximately 70% of observed paths. Network operators should treat ORIGIN rewriting as a significant deviation from BGP’s intended behavior, since it can alter routing decisions, distort traffic engineering, and create commercial incentives for further manipulation.

gitlab

GitLab and Anthropic: Governed AI for enterprise development (opens in new tab)

GitLab is expanding its integration with Anthropic Claude to provide enterprise teams with more capable AI inside a governed software development platform. Claude supports GitLab Duo Agent Platform features such as code generation, review, agentic chat, and vulnerability resolution. The central argument is that organizations should not have to trade advanced AI capabilities for security, compliance, and auditability. ## Governed AI across the SDLC - Claude-generated changes follow GitLab’s existing merge request process, approval rules, security scans, and audit trails. - AI agents do not bypass controls; their actions remain attributable, reviewable, and subject to policy enforcement. - This governance becomes increasingly important as agents autonomously plan, code, test, secure, and deploy software. - GitLab positions built-in governance as a core architectural differentiator rather than an added feature. ## Flexible enterprise deployment - Claude is available in GitLab through: - Google Cloud Vertex AI - Amazon Bedrock - Organizations can use existing cloud contracts, governance frameworks, and data-residency arrangements. - GitLab’s availability in the Claude Marketplace lets customers purchase GitLab Credits and apply them toward Anthropic spending commitments. - These options simplify procurement and consolidate AI spending. ## Supporting an agentic development model - GitLab is selecting model partners based on reasoning ability, reliability, and safety. - The platform is designed to maintain visibility into what AI agents do, when they act, and how their changes are tracked. - As agents take on more complex engineering tasks, GitLab argues that strong models must be paired with equally strong governance. ## Implications for customers - Existing GitLab Duo users gain deeper Claude-powered assistance without changing their established governance processes. - Organizations evaluating AI development platforms can access advanced models while retaining enterprise control. - GitLab presents the integration as a way to accelerate development without compromising compliance or oversight. The practical recommendation is to evaluate AI platforms not only by model capability, but also by how well they integrate governance, auditability, cloud deployment options, and existing enterprise workflows.

cloudflare

A closer look at a BGP anomaly in Venezuela (opens in new tab)

The post examines a January 2 BGP anomaly involving Venezuela’s CANTV network (AS8048). Although the event prompted speculation about government-directed surveillance, the broader pattern—eleven leaks since December—more strongly suggests inadequate BGP import and export filtering. The affected routes, provider-customer relationships, and heavy AS-path prepending are consistent with a configuration or operational error rather than clear malicious activity. ## Background: How BGP Route Leaks Work - BGP directs traffic between autonomous systems (ASes) according to business relationships: - **Customer-provider:** Providers advertise broad Internet routes to customers; customers advertise their own and downstream routes. - **Peer-peer:** Networks exchange their own and customer routes without payment. - These rules produce “valley-free” paths, where traffic generally moves from customers toward providers and then back down toward customers. - A **route leak**, formally defined in RFC 7908, occurs when routing announcements are propagated beyond their intended scope. - A typical leak happens when a customer receives routes from one provider and incorrectly advertises them to another provider, causing traffic to take an inefficient or overloaded path. ## The AS8048 Route Leak - Cloudflare Radar identified AS8048, operated by Venezuelan ISP CANTV, as the leaking network. - CANTV learned routes from AS6762, Italian telecom Sparkle, and redistributed them to AS52320, Colombia’s V.tal GlobeNet. - The leaked prefixes were originated by AS21980, Dayco Telecom, and belonged to the same `200.74.224.0/20` subnet. - Since AS8048 appears to be a provider for AS21980, the leak may reflect incorrect route export policies involving a customer’s prefixes. - The post emphasizes that route leaks are common and are usually caused by mistakes or weak routing controls rather than deliberate attacks. ## Evidence from Routing Relationships - Cloudflare Radar, bgp.tools, and BGPKIT data all indicate a provider-customer relationship between AS8048 and AS21980. - BGPKIT’s relationship analysis showed: - AS8048 was identified as the upstream provider in 9.4% of observations. - AS21980 was almost never identified as upstream. - Although only 9.9% of route collectors saw the two ASes as directly adjacent, the available paths strongly supported AS8048 being AS21980’s provider. ## Significance of AS-Path Prepending - Many leaked routes included repeated instances of AS8048 in their paths. - AS-path prepending is normally used to make a route less attractive and shift traffic away from a particular connection. - A path such as `52320,8048,8048,8048,...,21980` does not mean traffic physically traversed AS8048 repeatedly; the repeated entries are routing-policy padding. - The heavy prepending would have made the leaked routes less preferred, which weakens the case that AS8048 was intentionally trying to attract large volumes of traffic for interception. - The available evidence therefore points more toward poor routing configuration or operational practice than a purposeful man-in-the-middle operation. CANTV’s repeated route leaks and apparent lack of effective routing policies should still be treated as a serious reliability and security concern. However, the post’s evidence supports interpreting this incident as an example of recurring BGP misconfiguration unless further data demonstrates intentional manipulation.

figma

Meet the Makers Defining Tech’s Next Chapter | Figma Blog (opens in new tab)

The post previews Config 2025 and introduces speakers exploring how technology can become more human-centered. Their work spans AI-native devices, creator tools, AI interfaces, and expressive robotics. Together, they suggest that technology’s next chapter will depend not only on technical capability, but also on accessibility, empathy, and thoughtful design. ## Config 2025 and Its Vision - Figma’s Config conference will take place in San Francisco from May 6–8 and in London on May 14. - The event will examine how AI and automation are changing the way people create and experience technology. - Figma emphasizes tools and experiences that respond to evolving human needs. - In-person London tickets are sold out, while San Francisco tickets and free virtual attendance remain available. ## Building the Next Computing Platform - Andrew “Boz” Bosworth, Meta’s CTO and Head of Reality Labs, began coding through a 4-H club and later taught Mark Zuckerberg’s AI class at Harvard. - As Facebook’s 10th engineer, he helped create the News Feed. - He now leads Meta’s work on AR glasses, mixed-reality headsets, and the metaverse. - Bosworth envisions AI-native devices and proactive, personalized assistants becoming part of everyday computing. - At Config, he will discuss Meta’s efforts to develop a new computing platform. ## Making Creator Tools More Inclusive - Ebi Atawodi, Director of Product Management for YouTube Studio, combines experience in literature, technology, and product leadership. - Her work focuses on helping creators develop ideas and reach new audiences, including through AI-powered features. - Atawodi stresses that technology must be accessible and designed for people who have historically felt excluded by products. - Her approach centers on creating tools that serve a broad range of creators rather than assuming a single user experience. ## Designing Interfaces for Advanced AI - Joel Lewenstein, Head of Product Design at Anthropic, works on the interface for Claude. - He compares designing AI systems to watching a child discover new abilities, highlighting the uncertainty and possibility of emerging technology. - AI’s growing sophistication challenges traditional assumptions about how software interfaces should work. - Lewenstein argues that strong designers are defined more by how they think than by experience in a particular industry. ## Giving Robots Personality - Dr. Madeline Gannon uses her research studio, Atonaton, to transform industrial robots into expressive, lifelike mechanical creatures. - Supported by a Knight Foundation grant, she investigates how art and empathy can influence relationships with autonomous machines. - Her work explores how easily people attribute life and intention to nonliving objects. - Rather than reproducing the complexity of the human mind, Gannon demonstrates how movement, appearance, and context can make robots feel animated. Config’s featured makers represent a broad view of technological progress: more capable devices and systems must also be accessible, intuitive, expressive, and emotionally resonant. The conference aims to explore that balance through the experiences of people actively shaping AI, design, and robotics.