With a many lines of YAML, Calico will keep watch as you make operation- controlled networking.

One of the intriguing aspects of moving to a top-down, operation-centric way of working is redefining how we do networking. Important as the operation model first abstracted down physical structure with virtualization and is now using Kubernetes and analogous unity tools to epitome down the underpinning virtual machines, networking is moving down from general- purpose routed protocol heaps to software- driven networking that uses common protocols to apply operation-specific network functions.

We can see how networking is evolving with Windows Garçon 2022’s preface of SMB over QUIC as an volition to general- purpose VPNs for train sharing between on- demesne Azure Mound systems and the Azure public pall. Also, in Kubernetes, we ’re seeing technologies similar as service mesh give an operation- defined networking model that delivers network morass with your distributed operation as part of the operation description rather than as a network that an operation uses.

A new networking subcaste operation- defined networking

This operation- driven networking is a logical extension of important of the software- defined networking model that underpins the public pall. Still, rather of taking deep understanding of networking and, more importantly, network tackle, it’s a shift to a advanced- position approach where a network is automatically stationed using the intents in policy and rules. The shift down from both the virtual and the physical is essential when we ’re working with stoutly tone-orchestrating operations that gauge up and down on demand, with cases across multiple regions and topographies all part of the same operation.

It’s still early days for operation- driven networking, but we ’re seeing tools appear in Azure as part of its Kubernetes perpetration. One option is the Open Service Mesh, of course, but there’s another set of tools that helps manage the network security of our Kubernetes operations Network Policy. This helps manage connectivity between the colorful factors of a Kubernetes operation, handling business inflow between capsules.

Network programs in Azure Kubernetes Service

AKS (Azure Kubernetes Service) offers network policy support through two routes its own native tool or the community- developed Calico. This alternate option is maybe the most intriguing, as it gives you across-cloud tool that can work not only with AKS, but also with your own on- demesne Kubernetes, Red Hat’s Open Shift, and numerous other Kubernetes executions.

Calico is managed by Kubernetes security and operation company Tigera. It's an open source perpetration of the Kubernetes network policy specification, handling connectivity between workloads and administering security programs on those connections, adding its own extensions to the base Kubernetes functions. It’s designed to work using different data aeroplanes, from eBPF on Linux to Windows Host Networking. This approach makes it ideal for Azure, which offers Kubernetes support for both Linux and Windows holders.

Setting up network policy in AKS is important. By dereliction, all capsules can shoot data anywhere. Although this is n’t innately insecure, it does open up your cluster to the possibility of concession. Capsules containing back- end services are open to the outside world, allowing anyone to pierce your services. Enforcing a network policy allows you to insure that those back- end services are only accessible by frontal- end systems, reducing threat by controlling business.

Whether using the native service or Calico, AKS network programs are YAML documents that define the rules used to route business between capsules. You can make those programs part of the overall overload for your operation, defining your network with your operation description. This allows the network to gauge with the operation, adding or removing capsules as AKS responds to changes in cargo (or if you ’re using it with KEDA (Kubernetes- grounded Event- Driven Autoscaling), as your operation responds to events).

Using Calico in Azure Kubernetes Service

Choosing a network policy tool must be done at cluster creation; you ca n’t change the tool you ’re using once it’s been stationed. There are differences between the AKS native perpetration and its Calico support. Both apply the Kubernetes specification, and both run on Linux AKS clusters, but only Calico has support for Windows holders. It’s important to note that although Calico will work in AKS, there’s no sanctioned Azure support for Calico beyond the being community options.

Getting started with Calico in AKS is fairly simple. First, produce an AKS cluster and add the Azure Container Networking draw-in to your cluster. This can host either AKS network policy or Calico. Next, set up your virtual network with any subnets you plan to use. Once you have this in place, all you need to do is use the Azure command line to produce an AKS cluster, setting your network policy to “ calico” rather than “ azure.” This enables Calico support on both Linux and Windows not pools.However, make sure to register Calico support using the EnableAKSWindowsCalico point flag from the Azure CLI, If you’re using Windows.

The Calico platoon recommends installing the calicoctl operation tool in your cluster. There are several different options for installation running binaries under Windows or Linux or adding a Kubernetes cover to your cluster. This last option is presumably stylish for working with AKS as you can also mix and match Windows and Linux capsules in your cluster and manage both from the same Kubernetes terrain.

Structure and planting Calico network programs

You’ll produce Calico network programs using YAML, setting programs for capsules with specific places. These places are applied as cover markers when creating the cover, and your rules will need a chooser to attach your policy to the capsules that meet your app and part markers. Once you’ve created a policy, use kubectl to apply it to your cluster.

Rules are easy enough to define. You can set doorway programs for specific capsules to, say, only admit business from another set of capsules that match another chooser pattern. This way you can ensure your operation back end, say, only receives business from your frontal end, and that your data service only works when addressed by your aft end. The performing simple set of doorway rules ensures insulation between operation categories as part of your operation description. Other options allow you to define rules for namespaces as well as places, icing separation between the product and test capsules.

Calico gives you fine-granulated control over your operation network policy. You can manage anchorages, specific operation endpoints, protocols, and indeed IP performances. Your programs can be applied to a specific namespace or encyclopedically across your Kubernetes case. Rules are set for doorway and exit, allowing you to control the inflow of business in and out of your capsules, with programs denying all business piecemeal from what's specifically allowed. With Calico, there’s enough inflexibility to snappily make complex network security models with a sprinkle of simple YAML lines. Just produce the YAML you need and use calicoctl to apply your rules.

Operation-driven networking is an important concept that allows operation development brigades to control how their law interacts with the underpinning network fabric. Like storehouse and — thanks to tools like Kubernetes — cipher, the capability to treat networking as a fabric that can be simply controlled at a connection position is important. Networking brigades no longer have to configure operation networks; all they need to do is help define VNets and also leave the operation programs up to the operation.

Still, in ultramodern operations, we need to take advantage of tools similar to Calico, If we’re to make flexible. It may be a change in how we suppose about networks, but it’s an essential one to support ultramodern operation architectures.

Source

Unable to sign into your Microsoft Office 365, Azure Active Directory and other services? A global Multi-Factor Authentication (MFA) issue may be the reason.

A number of Microsoft Azure and Office 365 users have been unable to get into their accounts for most of the day on November 19. The problem: A multi-factor authentication issue which hit users worldwide and left them unable to sign into their services.

The Office 365 status page noted that affected users may be unable to sign in using multi-factor authentication (two-factor authentication) and may also be unable to do self-service password resets.

"A subset of users are no longer receiving prompts on their mobile devices (SMS, call or push) and (we) are investigating diagnostic logs to understand why," the status dashboard noted.

Azure and Office 365 MFA services were affected starting at 4:39 am UTC, which was 4:39 a.m. ET, according to the Office status page.

The Azure status page noted that those affected may have had problems signing into Azure resources like Azure Active Directory when MFA was required.

Azure engineers reported around noon ET today that they had deployed a hotfix, but that it took time to take effect across impacted reasons and in particular Europe and Asia-Pacific. Engineers were reporting they have seen a reduction in authentication errors as a result of the hotfix. The hotfix also tipped off engineers that some subset of customers were not receiving SMS, Call or Push prompts for MFA.

"Engineers are continuing to explore additional workstreams and potential impact on customers in other Azure regions to fully mitigate this issue," the Azure status page noted.

Beyond those status reports, Microsoft hasn't shared information as to what caused the MFA services problem today.

Update (4:30 pm ET): I'm hearing from more users that the MFA issue seems to be mostly over. The Azure status page now says all services are running normally. The Office 365 status page says:

"We've observed continued success with recent MFA requests and are taking additional actions in the environment in an effort to prevent new instances of this problem. Our investigation into the root cause of this problem is ongoing and maintained as our highest priority."


SOURCE:

Not all Microsoft 'partners' are created equal. These days, as in the case of Microsoft's latest 'partnership' announcement with Walmart, they are actually customers.

Microsoft and Walmart announced a "strategic partnership" back in July. At that time, Walmart committed to using Azure, Microsoft 365 and Microsoft AI and Internet of Things (IoT) tools and technologies to modernize its retail operations.

On November 5, Microsoft and Walmart officials said they were adding "an extension" to their original five-year deal. That extension involves Walmart expanding its existing Innovation Hub in Austin, Texas. The so-called Walmart "cloud factory," slated to open in 2019, will be staffed by 30 technologists, including engineers from both Walmart and Microsoft, according to a Microsoft blog post about the deal.

As part of this expansion, Walmart is planning to move thousands of its internal apps to Azure, plus build some new cloud-native applications. Walmart officials say they are going to use Microsoft's Cognitive Services, machine learning and chatbot technologies.

Microsoft's Walmart win is worth noting for a few reasons.

First, it's an example of the type of customer that Microsoft is targeting by positioning itself as an alternative to Amazon -- which is both a cloud vendor and a competitor to other brick-and-mortar retailers.

The Walmart deal also gives me an opportunity to talk about Microsoft's redefinition of the word "partner."

Until a couple of years ago, when Microsoft officials talked about partners, they meant either reseller/integrator partners or they meant OEM/ISV partners. More recently, Microsoft uses the partner/partnership terms a lot more loosely. Some Microsoft customers are now "partners," too.

Microsoft's justification in calling customers partners seems to be that many times -- as in Walmart's case -- Microsoft engineers and managers end up working side-by-side with the company's customers' engineering and management teams. This actually isn't new; Microsoft has been embedding its own engineers and support people inside key customers' shops for years, maybe decades.

But Microsoft officials maintain that things are different this time.

When Microsoft reorganized its salesforce a year ago and cut multiple thousands of jobs in the process, it also subsequently hired "more than 3,000 developers into the salesforce," according to Microsoft Executive Vice President of its Worldwide Commercial Business Judson Althoff. As a result, "we can actually code with our customers and help them really digitize everything they do within their business," Altoff said during the Citi Global Technology Conference in September.

As part of that reorg, Microsoft also built a "customer success organization," Altoff reminded attendees, which is basically a "non-billable consulting" organization, he explained.

"These are people that live with our customers beyond the deal to make sure that our cloud services actually get infused into business processes, that you don't have the cloud equivalent of shelfware in the software world. It's also been huge for us because the adoption and the actual utilization of the services have increased dramatically over the last year," he said.

(This is all part of Microsoft's big change in compensating salespeople for software/services sold vs. software/services that are actually consumed.)

Microsoft is going so far as to categorize some big customers officially as "digital partners," Altoff noted. At the Citi conference, he cited Boeing as an example, saying "we've actually taken their digital aviation assets, put them on our Azure platform, and we co-sell with them to other airlines around the world."

So for those trying to keep tabs at home, not all Microsoft "partners" are created equal. Some (many?) are actually customers, not traditional channel partners.


SOURCE:

Microsoft and NVIDIA are targeting data scientists, developers and researchers with preconfigured containers with GPU-accelerated software for running their AI and high-performance computing tasks.

1. NCv3 (1, 2 or 4 NVIDIA Tesla V100 GPUs)
2. NCv2 (1, 2 or 4 NVIDIA Tesla P100 GPUs)
3. ND (1, 2 or 4 NVIDIA Tesla P40 GPUs)

In a homecoming of sorts, cloud data warehouse pure-play Snowflake's product is no longer an AWS exclusive

Microsoft's service for bringing computing and artificial intelligence processing to IoT devices, Azure IoT Edge, is rolling out globally.

Axon reported strong fourth quarter results and detailed a December migration to Microsoft Azure as well as other cloud-centric implementations.

CoreOS's Kubernetes distro, Tectonic 1.7, conveys on half breed cloud by broadening compartment DevOps abilities crosswise over open-source and Azure mists and server farms.