Leaf-and-Spine Fabric Myths (Part 3)

Evil CCIE concluded his long list of leaf-and-spine fabric myths (more in part 1 and part 2) with a layer-2 fabric myth:

Layer 2 Fabrics can't be extended beyond 2 Spine switches. I had a long argument with a $vendor guys on this. They don't even count SPB as Layer 2 fabric and so forth.

The root cause of this myth is the lack of understanding of what layer-2, layer-3, bridging and routing means. You might want to revisit a few of my very old blog posts before moving on: part 1, part 2, what is switching, layer-3 switches and routers.

MUST READ: Operational Security Considerations for IPv6 Networks

A team of IPv6 security experts I highly respect (including my good friends Enno Rey, Eric Vyncke and Merike Kaeo) put together a lengthy document describing security considerations for IPv6 networks. The document is a 35-page overview of things you should know about IPv6 security, listing over a hundred relevant RFCs and other references.

No wonder enterprise IPv6 adoption is so slow – we managed to make a total mess.

Event-Driven Network Automation in Network Automation Online Course

Event-driven automation (changing network state and/or configuration based on events) is the holy grail of network automation. Imagine being able to change routing policies (or QoS settings, or security rules) based on changes in the network.

We were able to automate simple responses with on-box solutions like Embedded Event Manager (EEM) available on Cisco IOS for years; modern network automation tools allow you to build robust solutions that identify significant events from the noise generated by syslog messages, SNMP traps and recently streaming telemetry, and trigger centralized responses that can change the behavior of the whole network.

Why Is Network Automation such a Hot Topic?

This blog post was initially sent to subscribers of my SDN and Network Automation mailing list. Subscribe here.

One of my readers asked a very valid question when reading the Why Is Network Automation So Hard blog post:

Why was network automation 'invented' now? I have been working in the system development engineering for 13+ years and we have always used automation because we wanted to save time & effort for repeatable tasks.

He’s absolutely right. We had fully-automated ISP service in early 1990’s, and numerous service providers used network automation for decades.

Worth Reading: Software Disenchantment

Found an awesome blog post describing how we’re wasting resources on incomprehensible scale. Here’s a tiny little morsel:

Only in software, it’s fine if a program runs at 1% or even 0.01% of the possible performance. Everybody just seems to be ok with it. People are often even proud about how much inefficient it is, as in “why should we worry, computers are fast enough”.

Making Sense of Software-Defined World

In mid-September I was invited to present at the vNIC 2018 event in Frankfurt, Germany. Unfortunately I wasn’t able to get there, but Zoom did a great job … and enabled me to record the talk.

Worth Watching: Machine Learning in a Nutshell

This blog post was initially sent to the subscribers of my SDN and Network Automation mailing list. Subscribe here.

What could be better than an SDN product to bring you closer to a networking nirvana? You guessed it – an SDN product using machine learning.

Want to have some fun? The next time your beloved $vendor rep drops by trying to boost his bonus by persuading you to buy the next-generation machine-learning tool his company just released, invite him to watch James Mickens’ Usenix Security Symposium keynote with you.

Leaf-and-Spine Fabric Myths (Part 2)

The next set of Leaf-and-Spine Fabric Myths listed by Evil CCIE focused on BGP:

BGP is the best choice for leaf-and-spine fabrics.

I wrote about this particular one here. If you’re not a BGP guru don’t overcomplicate your network. OSPF, IS-IS, and EIGRP are good enough for most environments. Also, don’t ever turn BGP into RIP with AS-path length serving as hop count.

Feedback: Ansible for Networking Engineers

One of my subscribers sent me a nice email describing his struggles to master Ansible:

Some time ago I started to hear about Ansible as the new power tool for network engineer, my first reaction was “What the hell is this?” I searched the web and found many blah blahs about it… until I landed on your pages.

He found Ansible for Networking Engineers material sufficient to start an automation project:

VXLAN and EVPN on Hypervisor Hosts

One of my readers sent me a series of questions regarding a new cloud deployment where the cloud implementers want to run VXLAN and EVPN on the hypervisor hosts:

I am currently working on a leaf-and-spine VXLAN+ EVPN PoC. At the same time, the systems team in my company is working on building a Cloudstack platform and are insisting on using VXLAN on the compute node even to the point of using BGP for inter-VXLAN traffic on the nodes.

Using VXLAN (or GRE) encap/decap on the hypervisor hosts is nothing new. That’s how NSX and many OpenStack implementations work.

Worth Reading: The Fragile Engineers

Ethan Banks wrote an awesome blog post on the characteristics of fragile engineers (most of them probably being expert beginners). I can’t help but ponder how often I behave like one…

Leaf-and-Spine Fabric Myths (Part 1)

Apart from the “they have no clue what they’re talking about” observation, Evil CCIE left a long list of leaf-and-spine fabric myths he encountered in the wild in a comment on one of my blog posts. He started with:

Clos fabric (aka Leaf And Spine fabric) is a non-blocking fabric

That was obviously true in the days when Mr. Clos designed the voice switching solution that still bears his name. In the original Clos network every voice call would get a dedicated path across the fabric, and the number of voice calls supported by the fabric equaled the number of alternate end-to-end paths.

Network Automation Development Environments

Building the network automation lab environment seems to be one of the early showstoppers on everyone’s network automation journey. These resources might help you get started:

• I wrote an installation script that installs the myriad dependencies needed by Ansible and NAPALM in just the right order on a Ubuntu VM (step-by-step instructions for ipSpace.net subscribers).
• Carl Buchmann open-sourced a full-blown infrastructure-as-code development environment he uses for his automation projects.
• Jaap de Vos described how he creates a Docker image containing Ansible, NAPALM and Nornir.

Hint: after setting up your environment, you might want to enroll into the Spring 2019 network automation course ;)

Network Troubleshooting Guidelines

It all started with an interesting weird MLAG bugs discussion during our last Building Next-Generation Data Center online course. The discussion almost devolved into “when in doubt reload” yammering when Mark Horsfield stepped in saying “while that may be true, make sure to check and collect these things before reloading”.

I loved what he wrote so much that I asked him to turn it into a blog post… and he made it even better by expanding it into generic network troubleshooting guidelines. Enjoy!

Don’t Make a Total Mess When Dealing with Exceptions

A while ago I had the dubious “privilege” of observing how my “beloved” airline Adria Airways deals with exceptions. A third-party incoming flight was 2.5 hours late and in their infinite wisdom (most probably to avoid financial impact) they decided to delay a half-dozen outgoing flights for 20-30 minutes while waiting for the transfer passengers.

Not surprisingly, when that weird thingy landed and they started boarding the outgoing flights (now all at the same time), the result was a total mess with busses blocking each other (this same airline loves to avoid jet bridges).

Implications of Valley-Free Routing in Data Center Fabrics

As I explained in a previous blog post, most leaf-and-spine best-practices (as in: what to do if you have no clue) use BGP as the IGP routing protocol (regardless of whether it’s needed) with the same AS number shared across all spine switches to implement valley-free routing.

This design has an interesting consequence: when a link between a leaf and a spine switch fails, they can no longer communicate.

For example, when the link between L1 and C1 in the following diagram fails, there’s no connectivity between L1 and C1 as there’s no valley-free path between them.

Infrastructure-as-Code Tools

This is the fourth blog post in “thinking out loud while preparing Network Infrastructure as Code presentation for the network automation course” series. Previous posts: Network-Infrastructure-as-Code Is Nothing New, Adjusting System State and NETCONF versus REST API.

Dmitri Kalintsev sent me a nice description on how some popular Infrastructure-as-Code (IaC) tools solve the challenges I described in The CRUD Hell section of Infrastructure-as-Code, NETCONF and REST API blog post:

Upcoming Webinars and Events: October 2018

The fast pace of webinars continues in October 2018:

• Rachel Traylor will talk about graph theory and its relevance to reliable network design on October 8^th;
• The Amazon Web Services Networking webinar will start on October 11^th. The second session is planned for October 25^th;
• On October 16^th we’ll have the third session of VMware NSX technical deep dive (unless I manage to finish on time later today… not likely).

There are no on-site events planned until early December:

• We’ll run another on-site workshop in Zurich on December 5^th . This time we’ll focus on using VXLAN and EVPN to build multi-site fabrics;
• I’ll talk about making SDN better with IPv6 on December 6^th.

You can attend all upcoming webinars with an ipSpace.net webinar subscription. Online courses and on-site events require separate registration.

VXLAN Broadcast Domain Size Limitations

One of the attendees of my Building Next-Generation Data Center online course tried to figure out whether you can build larger broadcast domains with VXLAN than you could with VLANs. Here’s what he sent me:

I'm trying to understand differences or similarities between VLAN and VXLAN technologies in a view of (*cast) domain limitation.

There’s no difference between the two on the client-facing side. VXLAN is just an encapsulation technology and doesn’t change how bridging works at all (read also part 2 of that story).

Smart or Dumb NICs on Software Gone Wild

Hardware vendors are always making their silicon more complex and feature-rich. Is that a great idea or a disaster waiting to happen? We asked Luke Gorrie, the lead developer of Snabb Switch (an open-source user-land virtual switch written in Lua) about his opinions on the topic.

TL&DL version: Give me a dumb NIC, software can do everything else.

If you want to know more, listen to Episode 93 of Software Gone Wild.