Posts

Recently, a discussion on the internal Slack at Network to Code highlighted the “Amazon API Mandate” from 2002—a directive widely seen as transformative for Amazon. In the post (which can be found here), the core tenet is clear: All teams will henceforth expose their data and functionality through service interfaces. Upon rereading this mandate, I was immediately reminded of why I am such a strong believer in this strategy. Let’s take a deeper dive. First, the mandate included the following requirements:

1. All teams will henceforth expose their data and functionality through service interfaces.
2. Teams must communicate with each other through these interfaces.
3. There will be no other form of interprocess communication allowed: no direct linking, no direct reads of another team’s data store, no shared-memory model, no back-doors whatsoever. The only communication allowed is via service interface calls over the network.
4. It doesn’t matter what technology they use. HTTP, Corba, Pubsub, custom protocols — doesn’t matter.
5. All service interfaces, without exception, must be designed from the ground up to be externalizable. That is to say, the team must plan and design to be able to expose the interface to developers in the outside world. No exceptions.
6. Anyone who doesn’t do this will be fired.
7. Thank you; have a nice day!

Managing audio settings on Mac devices can be repetitive, especially when switching between different microphones for calls, recordings, or meetings. Siri Shortcuts provides a powerful way to automate this process.

AutoCon4 and NautoCon@AutoCon are in the books. It was quite the day, and I have to say that it ranks among my best days of the year in Network Automation — right up there with being able to meet up with the Network to Code team earlier in the year. It was great to see many friends from across the community. Hopefully we’ll be able to keep the conversation going in Slack and maintain the momentum.

It’s been a while since my last post — life and work have both been full. Today, I’m diving into one of the most quietly critical aspects of network design: time synchronization. Specifically, how to design NTP within an enterprise network and how to think about time zones when correlating logs.

We’ll look at two main topics in this post: designing NTP (Network Time Protocol) and managing time zone display in logs.

As the AI movement continues to expand its reach into the networking space, the need for an appropriate source of truth for network data becomes more critical than ever. What I have been seeing so far in the industry for networking and AI has been a lot of working on the individual devices one by one. But when looking at leveraging AI for the network, I believe it is best to look at the network as a whole. And that is where the Source of Truth data being stored in Nautobot is going to provide the right information about the network and the relationships between pieces of information - AI thrives on context, and relationships provide that context for more accurate insights and actions.

Coming out of the 2024 AutoCon2 conference held in Denver the week of November 18th, 2024 - there is one thing that is standing out more so than anything else. Containerlab is a HUGE blowout success. In observing through several of the workshops at the conference on Monday and Tuesday, many were using Containerlab in some fashion. Now, Containerlab has been around for a while, so this isn’t a press release of it. But it is re-affirming what many already know, that this is a great tool to be in the network engineer and network automator toolset.

This post is will provide a brief overview of how port binding works in Linux. This topic that will be required for the small series of using Continue.Dev in your local environment, but before addressing the setup of a machine for remote access in a future post, I thought it would be important to quickly create a post regarding the concept of port binding.

The year 2024 will showcase the remarkable evolution and contributions of AI. One prominent application of AI lies in its ability to streamline the coding process. In this post I demonstrate how to utilize the Continue VS Code plugin as a viable alternative to the GitHub Copilot system. This will allow you to have choice on the AI back end or in this scenario, the capability to self host the AI system using Ollama.

This may be my favorite post within the realm of what is possible as I write this series. There are many more things that could be done, but we did pretty well considering. Going back over a decade now, I had the fortunate opportunity to work to deploy Guest WiFi 📶 for a large number of retail sites, with the heavy lifting of the work being done within a five month period of time. This post is about the conversion of access points from one management system to another.

In the first release of Nautobot all of the Jobs were Atomic by default. This was from the previous focus of the legacy source application that assumed that scripts/reports would only be run on the data locally, so by that nature, the jobs should be atomic. As more and more Jobs started to interact with other systems, it became apparent that there needed to be a control mechanism provided (as I understand). So the introduction of a context manager and decorator was brought to the table to provide the same previous functionality while changing of the default behavior in Nautobot 2.x+.

At a previous position to joining Network to Code I was asked to help to build automation to help with the configuration of switches going into a MLS stadium. The stadium was under construction and the network build out would take place at the same time as the stadium was being built out. It was definitely a first and maybe only opportunity that I would have to build out a new stadium.

Scenario #

The task at hand is that each of the ports would need to be configured leveraging a good L2/L3 separation with each of the service providers that provide a service to the stadium their own network segment to work through. A large number of ports were going to need to be configured.

Nornir includes a function that allows for the transformation of inventory data, a feature integrated within the Nornir platform itself. The documentation for Nornir 3.0 is somewhat sparse regarding the usage of Transform functions, so I often refer to the more comprehensive 2.5 documentation. According to the Nornir documentation:

A transform function is a plugin that manipulates the inventory independently from the inventory plugin used. Useful to extend data using the environment, a secret store or similar.

In this post, we’ll dive into WAN design and address a common question that I was provided with in the 2000s: “My home internet costs only $35 per month. Why do we spend $xxx per month per circuit?”

I have been looking at migrating over to the Starship shell for a little while. The allure of running a rust shell prompt that gives me a ton of information is what I look for in a shell prompt. Such as the prompt below:

joshv in 🌐 my_device in nautobot on  u/jvanderaa-update_install_home_doc is 📦 v2.2.5b1 via 🐍 v3.11.9 (nautobot-py3.11)

The default installation however did not get the same behavior as my previous Oh My Zsh set up with the zsh-autocompletions and zsh-syntaxhighlighting. Whenever I would hit the up arrow key, the system would cycle through the commands as comes default with zsh/bash. But I was looking for subcommand scrolling. Such as the following command sequence.

So, if you are a returning user you notice something that is a little bit different. I have migrated my blog site to a new site generator and format. I am now using [mkdocs-material] blog them that was introduced in late 2022 to the platform. The migration was not too painful to make, but I’m feeling in a good place about it at this point. The blog has been on a journey so far. It started off with a Jekyll themed site which I liked. Then I decided to move onto something that has a little more development. So I moved into the land of Hugo.

Today I’m going to dive into my getting started with network automation, and perhaps my first successful automation. There are definitely some things that I would re-do and complete differently, and some things that I consider a success.

I’m working on a new series within my blog, about how I would look to have done things differently than I had done before, with the tooling and knowledge that I now have, years later. This is the first in the series.

One of my favorite features of Nautobot that may not be well known is the capability to put a button on pages that take you to other locations. This can be helpful when lining up the source of truth as that first place that you go, the idea of adding custom links will just help to enforce that as the first place to go. When you look at the idea of a source of truth to help feed other systems, you start to see the topology like below.

Just recently released at the beginning of 2024 is a project that I am super excited to see in the open source by Network to Code. This is the Nautobot App cookiecutter template. This may already be the biggest thing to become available for Network Automation in 2024. I know, its fresh at this point in the year, but this is something that is going to make getting started with your own Nautobot Application so much quicker.

The year of 2023 I think may have had some of the biggest leaps in the Network Automation capabilities that are being delivered by some of the best in the business. With Nautobot’s Golden Config App adding the ability to complete configuration remediation and Ansible release Event Driven Ansible, there are a couple of powerful tools to help you with your Network Automation. And all with a great new conference addition specific to Network Automation.

Recently Slack started introducing a new UI that is bringing back the importance of knowing keyboard shortcuts. Keyboard shortcuts within Slack are immensely helpful in my day to day, and it is essential for me being able to keep up with what is going on within my organization and some of the other 20 Slack organizations that I have joined. So much so, I don’t believe that once I am in the Slack UI on my machines, I am ever touching the mouse to get from one place to another. About the only time I may need to use the mouse is to scroll back in the thread, so I’m not even clicking when using the mouse.