This is the liveblog from the day 1 general session at VMworld 2019. This year the event is back at Moscone Center in San Francisco, and VMware has already released some juicy news (see here, here, and here) in advance of the keynote this morning, foreshadowing what Pat is expected to talk about.

The keynote kicks off with the usual inspirational video, this one incorporating themes and references from a number of high-tech movies, including “The Matrix” and “Inception,” among others. As the video concludes, Pat Gelsinger takes the stage promptly at 9am.

Gelsingers speaks briefly of his 7 years at VMware (this is his 8th VMworld), then jumps into the content of his presentation with the theme of this morning’s session: “Tech in the Age of Any”. Along those lines, Gelsinger talks about the diversity of the VMworld audience, welcomes the attendees in Klingon, and speaks very quickly to the Pivotal and Carbon Black acquisitions that were announced only a few days ago.

Shifting gears, Gelsinger talks about “digital life” and how that translates into millions of applications and billions of devices and billions of users. He talks about how 5G, Edge, and AI are going to further revolutionize digital life, and how that revolution will further affect the technology industry. Coming out of this discussion, Gelsinger waxes philosophical about the nature of technology, and the ability of technology to affect mankind for both good or ill. Gelsinger believes in “tech as a force for good,” and believes that VMware and the VMware community bear an enormous responsibility to make that a reality. Gelsinger brings Callum Eade, who swam the English Channel as a fundraising effort to help fight a deadly childhood cancer, as an example of working as a “force for good.”

Transitioning, Gelsinger talks about “the law of unintended consequences,” and how various technological advances and how they’ve had unintended and unexpected results. He talks about how VMware works to do “good engineering” and “engineering for good.” This leads to a video about Techsoup (a company that helps various nonprofits) and MotherCoders (a nonprofit focused on helping mothers break into the technology industry).

As a result of the impact of technology and the responsibility that technologists have to help ensure technology is used for mankind’s good, Gelsinger believes that there is no more important time to be a technologist. He addresses the audience and emphasizes that “you” (the audience) needs to bear this responsibility.

Transitioning yet again, Gelsinger now focuses on talking about VMware’s vision, which—in his words—remains unchanged in aiming to deliver any application to any device on any cloud. His discussion leads into a multi-cloud discussion and how VMware is “uniquely” (one of Gelsinger’s favorite words) positioned to help companies “build, run, and manage” applications across any cloud.

Starting with the “build” portion of the “build, run, manage,” narrative, Gelsinger makes his first mention of Kubernetes. Gelsinger believes Kubernetes will connect the developer and the operator, and he quotes Joe Beda in saying that Kubernetes is “improvisational jazz.” With that, Gelsinger brings Joe Beda—now a Principal Engineer at VMware—onto stage.

After some banter between Gelsinger and Beda, Gelsinger makes the official announcement of VMware Tanzu (pronounced “tahn-zoo” in spite of how Gelsinger said it on stage) as a suite of products around the “build, run, manage” narrative. The discussion following the Tanzu announcement also brings in the Bitnami and Pivotal acquisitions and how they will fit into the larger picture within the Tanzu framework.

Immediately on the heels of the Tanzu announcement, Gelsinger and Beda make the Project Pacific announcement. Pacific is all about embedding Kubernetes directly into vSphere. (Make no mistake—having seen the details from the inside of VMware, this isn’t a cosmetic integration. This is deep, deep integration.) Beda shares a few details on preliminary performance results showing Pacific running workloads 30% faster than Linux VMs and 8% faster than bare metal environments.

Moving on to the “manage” portion of the “build, run, manage” narrative, Gelsinger and Beda announce VMware Tanzu Mission Control, a cloud-neutral platform for managing Kubernetes clusters no matter where they are. That includes DIY Kubernetes on-premises, DIY Kubernetes on the public cloud, or managed Kubernetes offerings such as AKS, EKS, or GKE. No need to have these clusters running on vSphere; Tanzu Mission Control can manage those clusters where they are today.

Beda leaves the stage, and Gelsinger shifts to talk about CloudHealth. This leads into an announcements around CloudHealth Hybrid, the integration of SecureState into CloudHealth, and enhanced integration with the vRealize suite. Gelsinger takes this opportunity to talk briefly about the differences between multi-cloud (diverse resources and operational models) and hybrid cloud (consistent operational models).

Following on that discussion, Gelsinger talks for a few minutes about VMware Cloud Foundation, and that in turn leads into a discussion about VMware Cloud on AWS (which is built on VMware Cloud Foundation). Topics that Gelsinger touches on include following up on the AWS Outposts announcement, and expansions of VMware Cloud in AWS into new regions worldwide.

Gelsinger points out that using VMware Cloud on AWS helps customers avoid the costly “replatforming” efforts that are typically seen when modernizing applicatson to run on a public cloud platform. He reiterates that Tanzu will be available on VMware Cloud on AWS, and points to new partnerships such as one with NVidia to enable new types of workloads on VMware Cloud on AWS. To emphasize that last point, Gelsinger shows a video from the CEO of NVidia talking about how the VMware-NVidia partnership enables AI and ML developments.

Gelsinger announces the global availability of VMware Cloud with Microsoft Azure (I’m sure I’m violating some naming/branding stuff here), and also announces the availability of VMware Cloud on Dell EMC. Both are build on VMware Cloud Foundation, and the Dell EMC offering leverages VxRail.

What about hybrid cloud operations? Gelsinger spends a few minutes talking about how VMware is building out a portfolio of products and solutions to help with hybrid cloud operations. He also spends some time discussing solutions for edge and telco. He follows that discussion with a video of Sanjay Poonen talking with various folks from Verizon about the impact of 5G and other technological innovations.

Naturally, a discussion of 5G leads into a discussion on networking, and this results in Gelsinger focusing on VMware NSX and related networking acquisitions such as Velocloud (for SD-WAN) and Avi Networks (now NSX Advanced Load Balancer). Gelsinger next announces NSX Intelligence, a combination of vRealize Network Insight and NSX. It’s not clear whether NSX Intelligence leverages other recent acquisitions such as Veriflow.

At this point, Gelsinger brings Sanjay Poonen, VMware’s COO, to the stage to lead a discussion with some customers. Poonen is, as usual, a dynamic and energetic speaker. Poonen brings out Rathi Murthy and Tim Snyder from Gap and Freddie Mac, respectively. Poonen makes some small talk with Murthy and Snyder before getting into discussions about how these companies are leveraging technology to achieve their business objectives. Poonen does make sure, of course, to ask Murthy and Snyder about their thoughts on the Tanzu announcements. I do appreciate that Poonen asked both Murthy and Snyder about how VMware can improve, although I’m also quite aware that the discussion was probably heavily scripted.

After Poonen sends Murthy and Snyder off the stage, he launches into a discussion of digital workspace. Unfortunately, I did not capture any of Poonen’s announcements around digital workspace (and Workspace ONE, presumably) as I needed to leave the keynote for a meeting.

…

Read MoreLiveblog: VMworld 2019 Day 1 General Session

Welcome to Technology Short Take #118! Next week is VMworld US in San Francisco, CA, and I’ll be there live-blogging and meeting up with folks to discuss all things Kubernetes. If you’re going to be there, look me up! Otherwise, I leave you with this list of links and articles from around the Internet to keep you busy. Enjoy!

Networking

• Networking guru Ivan Pepelnjak has migrated his online presence to AWS; read more here.

Servers/Hardware

• Interesting (but otherwise not terribly useful) article on how to turn a MacBook into a touchscreen. Lack of a touch screen remains the MacBook line’s second most egregious shortcoming against competing products (the first being the awful keyboard).

Security

• I came across this article on tools and methods for auditing Kubernetes RBAC policies, which had some new tools I hadn’t seen before.
• This web developer security checklist has a nice list of some “current recommended practices” for properly securing web applications/sites.
• Daniele Antonioli, Nils Ole Tippenhauer, and Kasper Rasmussen discuss the KNOB attack (a Bluetooth vulnerability).

Cloud Computing/Cloud Management

• My teammate John Harris has a great post on ensuring least privilege in Kubernetes using impersonation. Good stuff here.
• Alex Brand (also a teammate!) has a post on using Sonobuoy to test Kubernetes Network Policy enforcement (this is to ensure that your installed CNI plugin has the abilitiy to enforce Netowrk Policies).
• I’ve been using Pulumi quite a bit to help with managing infrastructure environments (spinning up and tearing down test environments). I then found this article by Lee Briggs on using Pulumi for Kubernetes configuration management, and it has me interested to continue to explore Pulumi’s capabilities. (Oh, and while we’re talking about Pulumi, here’s another article on using Pulumi for infrastructure as code.)
• Gary Stafford has a fairly in-depth article on managing AWS infrastructure using Ansible, CloudFormation, and CodeBuild.

Operating Systems/Applications

• Here’s how to create a snippet in VS Code to wrap text with some other text of your choosing. This is handy—I had some similar things set up for Sublime Text, but haven’t managed to do the same for VS Code yet.
• Network bonding in Linux—as outlined in this article—is something I may have to explore. Right now I have a custom shell script I wrote that “docks” and “undocks” my laptop by manipulating, among other things, some associated network settings. What I’d really like to have is something like ControlPlane (née Marco Polo for the longtime Mac users), but for Linux. Alas, it appears that even ControlPlane has been abandoned…
• Major Hayden explains how to use Fedora 30 on GCE.
• jq, awk, and sed. Learn them, use them, love them.
• Ivan Velichko journals his journey of learning about containerization and orchestration in this blog post. If nothing else, this post at least helps clarify the relationship between commonly-mentioned projects like runC, containerD, cri-o, and others.
• Youichi Fujimoto discusses running Kubernetes locally with KinD and Docker. I’ve found KinD to be easier to use and more flexible than Minikube, but I’m on Linux and have a native Docker installation.
• Tim Little shares some “lessons learned” regarding increasing resilience in Kubernetes, such as taking advantage of features like Pod Anti-Affinity, scaling up Deployments to have more than a single Pod, and configuring Readiness probes for all workloads.

Storage

• Via this blog post by Myles Gray, VMware has announced Cloud Native Storage (CNS), powered by a new CSI (Container Storage Interface) driver for Kubernetes (and other orchestration platforms that use CSI).
• When it comes to storage, Cormac Hogan is an invaluable resource. It’s great to see Cormac turning his attention to Kubernetes environments, and in a recent post Cormac tackled failure scenarios with Kubernetes storage on vSphere.
• Laurens van Dujin ferrets out a small detail regarding network connectivity while troubleshooting vSAN Witness Node isolation.

Virtualization

• Joey Ketels steps readers through how to set up RHEL7 desktops with Horizon Instant Clones.

Career/Soft Skills

• Via the vExpert Slack instance, I came across this article on the effect of the scarcity mentality in IT careers.
• Silvia Botros has a great piece of how we keep learning. There’s some good stuff in here, I highly recommend reading it.
• Although written for the creative industry, I think this article by Paul Jun on five things no one tells you about going full-time again probably holds equally true for IT freelancers moving back into corporate life, or even moving from startup culture to big company culture after an acquisition.

I guess that’ll have to do for now. It is my sincere hope that you’ve found something useful in this post, and if you have any feedback I invite you to contact me on Twitter. Thanks for reading!

…

Read MoreTechnology Short Take 118

As I mentioned back in May in this post on creating a sandbox for learning Pulumi, I’ve started using Pulumi more and more of my infrastructure-as-code needs. I did switch from JavaScript to TypeScript (which I know compiles to JavaScript on the back-end, but the strong typing helps a new programmer like me). Recently I had a need to create some resources in AWS using Pulumi, and—for reasons I’ll explain shortly—many of the “canned” Pulumi examples didn’t cut it for my use case. In this post, I’ll share how I created tagged subnets across AWS availability zones (AZs) using Pulumi.

In this particular case, I was using Pulumi to create all the infrastructure necessary to spin up an AWS-integrated Kubernetes cluster. That included a new VPC, subnets in the different AZs for that region, an Internet gateway, route tables and route table associations, security groups, an ELB for the control plane, and EC2 instances. As I’ve outlined in my latest post on setting up an AWS-integrated Kubernetes 1.15 cluster using kubeadm, these resources on AWS require specific AWS tags to be assigned in order for the AWS cloud provider to work.

As I started working on this, I found examples of Pulumi TypeScript code that could do part of what I needed, but not all of it:

• If you’re using the awsx module with Pulumi (part of their “Crosswalk” stuff) to create a VPC, it can handle creating subnets across multiple AZs automatically for you. Users have the ability to fine-tune some of the settings, but—here’s the kicker—tags you specify at the VPC level don’t get propagated. (There’s an issue open for this.)
• If you build it all manually, then you also have to manually deal with figuring out how many AZs are in a region, and how to loop through the AZs to create subnets in each one (which is stuff the awsx module handles for you).

Since I needed the tags—the AWS cloud provider for Kubernetes won’t work otherwise—I set out to take the second path (building it all out manually). I’m going to share the TypeScript code I used—and am still using—so that others who may find themselves in the same boat have an example upon which to base their code.

(I’m a relative newcomer to TypeScript, so be gentle!)

First, I needed to figure out how many AZs were in the given region. This snippet of code returned both the list of AZ names as well as the total number of AZs in the region:

const rawAzInfo = aws.getAvailabilityZones({ state: "available",
});
let azNames: Array<string> = rawAzInfo.names;
let numberOfAZs: number = azNames.length;

I gathered both the list of AZ names as well as the overall count because in the snippet of code for creating subnets, I’ll need the AZ names.

Next, I created a VPC (this part is straightforward and reasonably well-documented):

const vpc = new aws.ec2.Vpc("k8s-vpc", { cidrBlock: "10.1.0.0/16", enableDnsHostnames: true, enableDnsSupport: true, tags: { Name: "k8s-vpc", [k8sTagName]: "shared", },
});

The one part here that wasn’t immediately obvious to me was the [k8sTagName] syntax for referencing a variable named k8sTagName. This allow me to declare the tag (i.e., “kubernetes.io/cluster/cluster-name”) once in a variable instead of multiple times in the code.

Now we get to the more interesting stuff—iterating through the list of AZs to create a subnet in each. After stumbling around for a bit, I finally arrived here:

let subnets = [];
for (let i = 0; i < numberOfAZs; i++) { let subnetAddr: number = i*16; let netAddr: string = "10.1."; let cidrSubnet: string = netAddr.concat(String(subnetAddr), ".0/20"); subnets.push(new aws.ec2.Subnet(`subnet-${i+1}`, { availabilityZone: azNames[i], cidrBlock: cidrSubnet, mapPublicIpOnLaunch: true, vpcId: vpc.id, tags: { Name: `subnet-${i+1}`, [k8sTagName]: "shared", }, }));
};

So what does this code do?

1. First, it creates an array to hold the aws.ec2.Subnet objects I’m going to create in the loop.
2. Next, it jumps into a for loop that iterates over the total number of AZs in the region (hence why I collected that value earlier).
3. For each iteration, the code calculates a subnet address (as a multiple of 16), and then combines that subnet address with the network address and CIDR mask. Thus, the CIDR block for the first subnet is “10.1.0.0/20”, the second is “10.1.16.0/20”, and so on.
4. Finally, for each iteration, the push method adds (“pushes”) a new aws.ec2.Subnet object into the array with the appropriate properties (the VPC ID for the VPC created earlier, the calculated CIDR block, any other settings, and the necessary tags). Note the subnet-${i+1} syntax to calculate names like “subnet-1”, “subnet-2”, etc.

After this, the majority of the rest of the code is pretty straightforward (I won’t go into detail in this post since the examples and Pulumi API reference pretty much capture most everything needed):

• Create an Internet gateway in the VPC and tag it.
• Create a route table for the Internet gateway and tag it.
• Use a similar array and for loop construct to iterate across the subnets and associate each subnet (referenceable by subnets[i], assuming i is the loop variable) with the route table.

At this point, Pulumi has now created a VPC, subnets in each AZ, an Internet gateway, a route table to point to the Internet gateway, and a route table association linking each subnet to the route table. All you need now are security groups and EC2 instances, and you’re off to the races! All of the things that can be tagged are tagged, which means infrastructure provisioned with this code is ready for use with Kubernetes.

It’s more than fair to say that some of this is just me learning TypeScript; I can’t dispute that at all. If I’m searching for resources on how to do something, though, that’s probably a reasonable indicator that other folks are also looking for resources and examples, and why I wanted to publish this post.

I hope that this information is useful to someone; feel free to hit me up on Twitter if you have any questions, comments, corrections, or suggestions.

…

Read MoreCreating Tagged Subnets Across AWS AZs Using Pulumi