netlab release 1.3.1 contains major additions to FRR and Cumulus Linux, and new BGP features:

• VXLAN, VLANs, VRFs, and EVPN implemented on FRR and Cumulus Linux
• BGP local-as implemented in the BGP configuration module and supported on Arista EOS, Cisco IOS, Dell OS10, FRR, and Nokia SR Linux.
• Configurable BGP transport sessions
• Configurable default BGP address families supported on Arista EOS, Cisco IOS, Cumulus Linux, FRR, and Nokia SR Linux.

Here are some of the other goodies included in this release:

The last video in the 2-hour-long Network Addressing part of How Networks Really Work discusses Network Address Translation.

After watching it, you might want to spend some extra quality time (with a bit of soap opera vibe) enjoying the recent ‌Dual ISP deployment operational issues and uncertainties thread on the v6ops mailing list with a “surprising” result: NPTv6 or NAT66 is the least horrible way to do it.

I keep hearing numerous variations of the following argument from people believing in the unlimited powers of multi-cloud¹ (deploying your workloads in multiple public cloud providers):

We don’t install all our servers in the same DC. But would you trust one Cloud Server Provider with all your applications? That’s why you should use multi-cloud.

I’ve been hearing similar arguments for at least 30 years, including:

Early bridges implemented a single bridging domain across all ports. Within a few years, we got multiple bridging domains within a single device (including bridging implementation in Cisco IOS). The capability to have multiple bridging domains stretched across several devices was still missing… until the modern-day Pandora opened the VLAN box and forever swamped us in the complexities of large-scale bridging.

One of my readers preparing for public cloud deployment sent me an interesting observation:

I pushed to use infrastructure-as-code as we move to Azure, but I’m receiving a lot of pushback due to most of the involved parties not having any experience with code. Management is scared to use any kind of “homegrown” tools that only a few would understand. I feel like I’m stuck deploying and managing the environment manually.

It looks like a bad case of suboptimal terminology for this particular audience. For whatever reason, some infrastructure engineers prefer to stay as far away from programming as possible¹, and infrastructure-as-code sounds like programming to them.

In the last blog post in the VLANs and VRFs in netlab series, I described how we can combine VLANs and VRFs and create a VRF Lite solution with stretched VLANs. Wonder how hard would it be to create a routed multi-hop VRF Lite topology? It’s trivial.

Andrea Dainese released an interesting tool that performs automated network discovery, pushes the discovered data into NetBox, and then uses netbox-topology-views plugin to create network topology diagrams.

Definitely worth exploring!

A recent blog post by Andrew Lerner asks whether Cisco ACI is dead. According to Betteridge’s law of headlines, the answer is NO (which is also Andrew’s conclusion), but I liked this gem:

However, Gartner assesses that Nexus Dashboard Fabric Controller is the optimal fabric management software for most Cisco data center environments.

An automation intent-based system provisioning a traditional routed network is considered a better solution than a black-box proprietary software-defined blob of complexity? Who would have thought…

After discussing rogue IPv6 RA challenges and the million ways one can circumvent IPv6 RA guard with IPv6 extension headers, Christopher Werny focused on practical aspects of this thorny topic: how can we test IPv6 RA Guard implementations and how good are they?

Long long time ago, Daniel Dib started an interesting Twitter discussion with this seemingly simple question:

How does a switch/router know from the bits it has received which layer each bit belongs to? Assume a switch received 01010101, how would it know which bits belong to the data link layer, which to the network layer and so on.

As is often the case, Peter Paluch provided an excellent answer in a Twitter thread, and allowed me to save it for posterity.

Network terminology was easy in the 1980s: bridges forwarded frames between Ethernet segments based on MAC addresses, and routers forwarded network layer packets between network segments. That nirvana couldn’t last long; eventually, a big enough customer told Cisco: “I don’t want to buy another box if I already have your too-expensive router. I want your router to be a bridge.”

Turning a router into a bridge is easier than going the other way round¹: add MAC table and dynamic MAC learning, and spend an evening implementing STP.

A few weeks ago, Daniel Dib tweeted a slide from Radia Perlman’s presentation in which she claimed IPv6 was the worst decision ever as we could have adopted CLNP in 1992. I had similar thoughts on the topic a few years ago, and over tons of discussions, blog posts, and creating the How Networks Really Work webinar slowly realized it wouldn’t have mattered.

netlab release 1.3 contains two major additions:

• VXLAN transport using static ingress replication or EVPN control plane – implemented on Arista EOS, Cisco Nexus OS, Dell OS10, Nokia SR Linux and VyOS.
• EVPN control plane supporting VXLAN transport, VLAN bridging, VLAN-aware bundles, and symmetric IRB – implemented on Arista EOS, Dell OS10, Nokia SR Linux, Nokia SR OS (control plane), VyOS, and FRR (control plane).

Here are some of the other goodies included in this release:

Etienne-Victor Depasquale, a researcher at University of Malta, is trying to figure out what technologies service providers use to build real-life metro-area networks, and what services they offer on top of that infrastructure.

If you happen to be involved with a metro area network, he’d love to hear from you – please fill in this survey – and he promised that he’ll share the results of the survey with the participants.

It’s so refreshing to find someone who understands the impact of latency on application performance, and develops a methodology that considers latency when migrating a workload into a public cloud: Adding latency: one step, two step, oops by Lawrence Jones.