As we head into the holidays, many of us are going to have quiet time to be with our families, kick back… and mess about with dev projects we’ve been wanting to get to for months.
For me, I’ll be busy diving into cloud native networking. I want to learn more about how to bridge the fields of compute and networking. I believe that bringing the two realms together leads to better network design and far greater scalability, cost efficiency, and reliability.
From a practical perspective, this means spending more time with Kubernetes and cloud native with CI/CD tooling.
My colleagues are getting even deeper into realms of cloud development. Here’s what several of them told me they’d be working on (playing with) — when they’re not stuffing themselves with holiday sweets.
Jason Davis: Building a microk8s cluster with MetalLB
My holiday project is focused on cloud native technologies and the application to traditional infrastructure networking. In particular, I’m looking at the challenges of exposing Kubernetes services publicly.
There are many technologies in this area – e.g., port-forwarding, LoadBalancer, ClusterIp, NodeIP, and Ingress Controller. I will be building a microk8s cluster that will use MetalLB to expose my services. It will be shared via BGP to Cisco routers for public access.
Quinn Snyder: Atlantis for Terraform
I’m fascinated with device compliance and sources of truth. Enterprises need both sources of truth for network configurations, and a low barrier to using them to make sure that no one goes rogue.
So I’ve been learning about GitOps and CD-centric pipeline applications, such as Atlantis. Atlantis integrates with both on-prem and cloud software configuration management (SCM) tools to provide a deployment workflow through pull requests. By adding/making a change to a Terraform configuration in a branch and making a pull request, Atlantis will gather the Terraform configuration, run a `terraform plan` against the target infrastructure, and place the results within the pull request. You tell Atlantis to apply the config using `atlantis apply`, and it will configure the devices for you using the Terraform plan output! Best of all, all configurations, changes, and PRs are stored using native SCM tooling as the source of truth. Automating deployment and compliance, without setting up tricky pipelines. It’s a win-win, and I’m looking forward to learning how best to apply this tool.
Pat Janakiram: Cisco Intersight Cloud Orchestrator
Lately, I’ve been spending a lot of time focusing on low code/no code ways to manage cloud infrastructure and to orchestrate network architectures. It’s taking me back to some of my favorite holiday memories: like playing with Legos for hours and hours – putting simple, colorful blocks together and creating complexity and order from them. Decades later, I am still building complex systems from simple blocks.
My holiday project is working with workflow designers. These are tools that break down complex IT tasks into simpler steps. They provide insights to help you find the weakest link in a job, and also help you prepare for future manipulation of workflows to meet changing business requirements.
Cisco has its own workflow service for deploying infrastructure and applications. It’s called Intersight Cloud Orchestrator (ICO), and it has helped me get started with workflow design. Here are a couple links where you can see some of the blocks I’ve started building, and keep an eye out as more content, workflows, and integrations:
- Step up your legacy Tomcat App Deployment with ICO and include insights with AppDynamics
- Improve Cloud Native Deployments with ICO and include insights with AppDynamics
You can use these library workflows as building blocks for more complex scenarios, swapping blocks from your library as you need!
Christopher van der Made: Bandit for Python
During this holiday, I will be working on security. And I’m excited about it! I will be taking a good look at some sample Python scripts that I am producing. Are they secure? I’m pretty sure they’re not. Therefore, I will be checking out multiple open source tools that can help with writing and deploying Python code into production, especially Bandit, a tool designed to find common security issues in Python code.
Bandit can be run from the command line to scan Python code. But to automate it, I will use GitLab to incorporate the Bandit security tests as part of a CI/CD pipeline. This way my code will be scanned before being deployed! What do I hope to get out of this? A more secure 2022!
Jock Reed: APIClarity, Harvester, and Karpenter
This holiday season, I am working on some cloud native, hyper-converged infrastructure (HCI) with Harvester. Harvester makes it easier to manage your infrastructure, and to lower total operating costs. It is built on top of Kubernetes and other CNCF projects for bare metal data centers. It’s a fairly new project put out by Suse, and it’s super cool. I will be sharing a lot of what I am doing with it in the new year – like spinning up developer-related workloads, running containers in groups alongside VMs, and a new take on cloud native, on-prem data center management.
I am also working on a very new open source project from AWS called Karpenter. It is a different take on Kubernetes cluster autoscaler, and should provide an easier way to scale cluster nodes. I am very excited about this one as it allows a lot of flexibility, and reduces the amount of work necessary to spin up new nodes in Kubernetes.
I have also been working a lot with a new cloud native project from Cisco called APIClarity. It’s an API tool that runs in tandem with service mesh to capture your API endpoint traffic and give you visibility into how your API microservices are being used.
Florian Pachinger: Apache Kafka
With the holidays coming up, I finally have some time to focus on a small new software project with Apache Kafka. I will be setting up Apache Kafka and streaming telemetry data from various sources to a time-series database (most probably InfluxDB).
For the data-generating sources I will stream telemetry data from Cisco IOS XE (with model-driven telemetry), from IoT sensor data, and from Thousand Eyes data. I might also try out an “edge scenario” where data will be streamed from Cisco’s rugged edge routers and the edge compute framework, IOx.
May this holiday season bring enormous happiness and peace to your life. See you all in 2022!
Have a comment or question?
Please leave me a note in the comments section below and follow me @bigevilbeard
We’d love to hear what you think. Ask a question or leave a comment below.
And stay connected with Cisco DevNet on social!
Visit the new Developer Video Channel