 Hello, and welcome back to another episode of AZ Update. I'm Anthony Bartolo, joined as always by producer Pierre. Pierre, how's it going? I'm good, how are you? Good, good. Hey, the Canadians came back and tied it up with Vegas 1-1. That is awesome. It's awesome that, you know, I guess all the candidates cheering for the Canadians now, which is amazing. Help us, Obi-Wan. You are our last hope. Seeing Tower lit up as the Canadian colors and all the Leaf fans got upset. But, you know, it's been an interesting go. Also been enjoying Euro, Euro Cup this week as well and last week and Copa America. So a lot of soccer on the TV and dawning. We had a lot of heat on the DM saying, why are you wearing Italian jerseys if you're Maltese? And I said, okay. So I donned my Maltese jersey. Had to pull it out and wear it. I know Maltese never made it to the World Cup or Euro Cup or they have and it's been a long time. I think since the last time they made it at least won the Stanley Cup. But you know what? Proud to be Maltese and definitely gone to the jersey. Producer Pierre. Yes. If a VM is not running, but it is in Azure, are you being built? You are not, if it's not running. Right. So is it shut down in the OS or is it shut down in the portal? So this is the interesting question that Sonia actually brought up this week. And she actually wrote a blog post about it. And she went through in detail the aspects of this question because this is one of the more popular questions that we receive, our team receives in regards to, hey, if a VM isn't running, but what does that mean that it's not running? Does it mean that it's been shut down via PowerShell, shut down through Azure CLI, shut down inside of the virtual machine itself? There's so many things that taking consideration, right? It's fairly simple. If it's been decommissioned or deprovisioned, which means the memory, the CPU, the networking has been, the resources have been released, then you're not paying for the compute and the network. If you shut it down in the OS, but you don't deprovision it, then you're still paying because those resources are earmarked or reserved for you. Because basically, it expects the machine to turn back on. That's what happens when you, after a patch or something and you reboot, you don't decommission the machine, you just shut the OS down and then turn it back on again. However, you're always paying for the storage for those disks that are hosting your VM. And that's the thing, right? You got to keep in consideration the different ways to shut down the VM itself. If you're shutting it down internally in the OS itself and doing the shutdown mode, like you said, there's still provisioning that, hey, this is going to be turned back on. So don't shut down all the services. So it depends, right? It depends on what your shutdown mechanism is. I know you actually crafted a PowerShell script that you would use to shut down lab VM so that we wouldn't incur major costs. As you said, there's still a cost incurred in terms of storage. Yeah, well, in the portal itself, when you actually create the machine, it'll ask you, do you want to enable auto shutdown? And then ask you what times you want to shut it down, do you want to be notified? Because when it notifies half an hour, you could say, oh, no, shut it down, I'm using it right now. Or you could use Azure Automation, which is something I do. In the PowerShell gallery, there's a number of PowerShell scripts that you can shut down the specific machine or reboot a specific machine or restart it or do whatever you need to. I had written a PowerShell script that would look in my subscription because my test description, and there's a good little story about that, where it would enumerate all of the VMs in my subscription and then go shut them down one by one. And the reason I'm doing that, I do that now is because during, and I've used this story before, especially when we're talking about cost management and policies and governance. In a demo, I was creating a VM scale set and I picked, I think it was a G5 size machine just to show, because it was fun to show, hey, there you go, 32 cores and four terabytes of memory. Ooh, it's really cool to show. But then at the end of my demo, I shut the lid off my laptop and went, thank you very much. Have a great day. And I had just deployed 10 G5 machines, which at the time, I think were like $8,000 a month. That was a while ago. Right. I remember that conversation. So at the end of the month, I had 10 machines times $8,000. I had an $80,000 bill that got charged to our cost center and my boss was like, what is that? So now I have this little automation that runs every night at 11.30 Eastern time and just enumerates all the VMs in my subscription, my test subscription and my demo subscription, sorry, and shuts them down. Then I have another one that runs. I had another one and I turned it on. I enable it when we're like doing like MITT tour, like MITT tour or something like that, where I'll turn them back on at eight o'clock in the morning. Right. But right now it's shut down. Regular IT guy in the chat says, I remember that expense report. Yeah, that didn't go well. So good morning to everybody in the chat. We have Jared Shockley here. We have regular IT guy, aka Rick Claus. We have Andrew McCollum and we have Robert Jr. Talking about, oh, emotional support email from the HBO Max intern email that came out yesterday. I got that same email as well. Spam is making the rounds. I'm surprised we didn't get more emails like that during the whole lockdown piece and everybody watching content online. I think with everybody now being able to go outside, we're gonna see a little bit less of that, but it's interesting to see that email come through. I received it. Did you get one of those Pierre? No, I don't, and I'm not familiar maybe. I was disconnected last night and this morning. So what's the story behind that? There's a spam email that's come out. It says it's from HBO Max. I didn't verify. I literally just saw it delete because I don't have the service, but it's just something that's interesting how it caught wind on social in terms of this email that's been going out and it's being shared out. So just keep an eye for that. I did see something on Twitter and it looked like somebody was testing a notification system and then the notification system actually sent it out to all of the subscribers because in Twitter, the screenshot actually says this is a template of what would go out and everybody was dumping on the DevOps person who set that up, which I think is a good thing. We shouldn't be jumping down the throats of people that are having issues, especially when the service is still up, it's just a monitoring problem. Let's jump on to the news. First up, there's apparently updates on Azure Monitor agent that you wanted to talk about. Yes, so, and I will just put up this little banner here to talk about Azure Monitor agent. If you see the article, it's been a while coming when you're doing monitoring, there is a plethora of agents. Yeah, there's one agent for your metrics. There's one agent for your logs. There's one agent for OMS. There's multiple agents. So every time you enroll a virtual machine in especially the ones on-prem for Azure Monitor or log analytics or metrics, you have to start deploying agents and configuring all those agents to send, and yet they can only send to one workspace. Now that's changing. The product group has figured out that this was not optimal. So now that we have one agent, the Azure Monitor agent, and what it does is it reads the data collection rules from the log analytics or Azure Monitor and then collects based on those rules and sends it to wherever it needs to go. So the deployment's a lot simpler because you only deploy one agent, configuring it to send the type of data to the right places is a lot simpler because you create rules and you create multiple rules and assign them to the same agent so you can have, for the people that want to have the hub and spoke Azure Monitor workspace set up, they could have the agent send the data to the hub and they can have the agent send the thing to the spoke so that for environment where there's multiple levels of management, keep in mind if you're doing that, it'll cost you double because you're sending it, you're ingesting that data twice. But there's a lot more flexibility in the deployment and the management of those agent. So now this is GA, generally available, which means if you're looking at onboarding workloads or workstations or virtual machines into Azure Monitor, don't look at the old agents, look at the new agents, so much simpler and better for ingesting data. And I'm guessing that it's collecting data from all facets. So even, there was the announcement of Azure Monitor agent forward at the edge. So IoT edge computing, it doesn't matter now, it's just that agent is listening in on the data stream as it comes in and captures that information. I'm not sure about the one at the edge because the agent that I know that they're talking about here is the one that runs on VMs for Windows and Linux, but it doesn't say anything about IoT gear. So if it's IoT that's Linux based or Windows based, then yes, it would work. But the way it works is the agent gets deployed and it just keeps polling for rules based on the identity of the VM. So you no longer have to put in the agent, the key and the GUID for the workspace where the agent is gonna be sending the stuff. It says, hey, I'm sitting on VM 001. I look to Azure and I say, okay, VM 01, which one, which DCR or data collection rule is assigned to me? This one, this one, this one. So I get it to collect my metrics from here and send them here, my logs from here and send them here, or line the business custom logs from here and send them here. And then it just collects it and sends it. Of course, it compresses it, encrypts it and sends it, but it's doing it all based on those data collection rules, which is a lot simpler to manage going forward than it was with the old agents. Plus you only have to manage one. Now you recently sat down with the product group into have this discussion, actually it was not even a month ago in terms of the updates that are coming out. And I think you had me mentioned here on this video that it was in preview. And so it's exciting now it's become generally available. Yeah, so if you go to that post and I'll put the URL again, if you go to that post in the interview, she actually gives demos on how you create the data collection rules and how it affects and applies to different machines. So this is a really good review. Even though this was in preview, it's still very, very relevant for everyone. And so this video and of course our AZ Update blog post that details all the news that we're talking about today available on itopstalk.com or go to AZ Update Show, aka.msforge slash AZ Update Show. And you get all the articles that we're talking about and the news that we're talking about as well as previous shows as well. Next up in terms of the news, Azure IoT Central updates made available. This is just launched for me. I know we're in June, usually the way the cycle works is that it finishes the end of the month and then we launch it in the next month. The big news here is the digital twin defined language made available through the JSON editor right in Azure itself. What does that mean? So when you're setting up your digital twin environment, you now have the capability of having the JSON file duplicated for your digital twin, digital twin being a replica of your existing on-premises IoT implementation in a virtualized sense so that you can test out, develop upon, you name it. Now the availability to actually see the JSON syntax directly in the JSON editor itself as part of the Azure IoT Central tools. This is important because if you're making changes, you don't have to make the changes and then deploy one by one or you can make the changes and deploy in real time in the digital twin, not affecting the in-production, on-premises implementation. This is from a perspective of ensuring that you even have version management specific to the digital twin and the ability to deploy that. It's a great opportunity to do so directly from the portal as opposed to the requirement of third party tools. I have to admit that I'm a noob when it comes to IoT. So if I understand correctly, digital twins is you have an IoT device and a configuration of a specific workload working on it and you have kind of like a virtual clone of that one sitting in the cloud so that you can test stuff in year before you screw up something on-prem. Correct, so that's the thing, right? You have a virtualized instance of your IoT implementation so that when you're developing solutions, you're developing it in the virtual plane. You're not doing it on production, but you're using the same code. And what's even cooler is the information that's coming in from the on-premises in-production implementation is being funneled into the virtualized plane so that you're always having this live data, even though it's virtualized, which is awesome. Regular IT guys made mention, I'm glad Anthony didn't say on-premise. There's a plethora of times you've told me regular IT guy not to say on-premise and now it's been drilled into my head so we're no longer saying on-premise unless it means well to say it. I've been caught too many times. That's why I now say on-prem. I just say on-prem, everyone, every time. Next up on the news, new cloud regions. Yes. This is a fun one because we open regions all the time. And just as Mark Rozinovich talked about, was it ignite or build? I can't remember. Any new region coming online is automatically configured with availability zones, which is great. But this one, not only does it have all the availability zones built in at GA, it's also self-sustainable. Right. So there's a picture online, and I don't know if you have it, of fields of solar panels. And of course, Twitter going nuts, going, oh, does that mean they shut down at night? No, there's batteries, there's other kinds of... But it is self-sustainable, and it is in our quest to become carbon neutral or carbon negative. So this is going to be cool. I think it's going to be West US 3, and it is online. So if you want to start running your workloads in that region to take advantage of the carbon footprint offset, you can use the Azure, is it called the Azure Resource Mover, and move some of your resource into that region. So that's really, really cool. And so my understanding too is that the region also is including the Azure Availability Zones, which they've mentioned before, when they create the new regions, this will be automatically included in all new regions to ensure resiliency. Yeah, because that's the thing is Availability Zones is basically three separate... It's basically three separate area that are completely independent from one to the other within a data center or within a region. When we, the first, let's say 50 regions that we released, didn't all have it, and then we started backfilling, so trying to turn them on as required. And now it's in part of the GA or the process of spinning up a new region it has to have Availability Zones. And that's awesome that they're doing it on all the new regions and they're also doing it on all existing regions. So I think the hope is that by the end of 2022, all regions that Microsoft deploys in Azure will include Availability Zones for that resiliency, which is important. Yeah, and I think we talked about it last week where some, we now have Availability of having Availability Zones across continent just with region pairs. Right. Yeah, it's exciting. Anything we can do to increase the resiliency of our service, which in turns, increase the resiliency of our customers' services is something that we strive to do. So let's continue on with the news. In the Northern Hemisphere, we're coming to summer and so that means that school is coming to an end and it's interesting because this article just came out this week in regards to preparation for next year. There was a lot of learning over the last 16 months especially because a lot of kids, a lot of students had to learn from home and so there was a lot of advancement in terms of technical adoption and technology adoption amidst the schools and the challenge is being, not every child has a computer and so a lot of schools sent out computers, a lot of students learned on tablets but there's also the back-end challenges as well in regards to just setting up the system to enable these classrooms to actually occur and to understand the resources that are being utilized in regards to this. So this great article that came out and again, we have this link on itobstock.com on the latest AZ Update Post talks about 20 updates that were made available specifically for schools IT administrators should take into consideration. I know in regards to, it's now summertime, everybody needs a break, 100% if you have that opportunity to do so, take it because your mental wellbeing is very important but these are a couple of things to take in consideration when you're planning out for the next school year. Couple of things I wanted to highlight, one here being the school data sync piece where it's taking in consideration your resources of what's being utilized from a school setting, resources in terms of whiteboard utilization or even in some instances virtual machines that are running labs and I've seen this, there's these virtual machines that are running math labs that are being spun up by teachers and then assigned to students. And so there's resources that are being utilized, who's using these resources at what times are these resources being used? How many classrooms are using the resources? Something even as teams is being used in classroom for that communication. That's all being tracked and then the insights are being brought forward through data analytics to understand, okay, so here's where your bottlenecks lie and these are the more compute power is required or more storage power is required from that aspect. And that's something that's great that you have through the school data sync to have that insights in terms of what is actually going on and it's your architecture. The other item I wanted to bring up really quickly because there was a lot, there was third party integration, there's the API management through Microsoft Graph that's also introduced. This one is a big one. Again, with all the third party applications, especially in schools of what's being utilized and every school has their own instance of what they use for learning. Microsoft has gone forth into M365 to create a centralized portal for management of your architecture, of your students and faculty members utilization. Here's an example in terms of teams and groups and the management of that capability specifically setting up channels for classrooms itself. This is a big thing in regards to, we talk about this a lot, centralized pane of glass to go through and understand your management requirements for the schools that you support, all done through the GUI interface on M365. Pierre, what are your thoughts? My thoughts is that is this is fantastic. My wife is a teacher and she's had to pivot to doing the things online and remote learning and stuff like that. So having these processes and these updates to facilitate that makes a huge amount of difference in her life and in my life, because I don't have to be the IT tech support for the school board because of course, when she's sitting at home and she has a problem with the computer, she calls us the IT guys there and they're just like, okay, well, it's not our machine, so we don't care. But having this all done in portals that are independent to the local setup of each of the teacher and the students makes a world of difference. In my area, the schools are set up and as a region as opposed to IT per school. So they have one IT service or one IT department that handles a plethora of schools in a specific geo. And so this would be an awesome tool for them to utilize because they would do this per school and then even get even more granular per class just from a management tool. So I hope, maybe they're looking into this, if not, hey, reach out, definitely interested and help it out if that's required. Let's do a quick shout out to those in the chat room because we've got more people who have joined us. So Andrew has it going. Henry Morales, I apologize. How's it going? Hello, you guys, Steve, Jared Chalkley has joined us. You got it right. You got it right. I've been practicing. It's like that in our premises. I've been practicing both. Speaking of on-premises, regular IT guy. Rabbi Jr. joins us. Paul Jensen, how's it going, sir? He's selling us the drink because we're using the word plethora too many times. And to everybody who's talking in the chat room, awesome for you to be here. We love the fact that you're interacting and chatting with us. And if you're watching this as a recording, definitely check us out on live stream and join into the chat because we always get great ideas and great questions that come into the chat room as well. Let's start talking about it. And if you're watching this offline, if you're watching the recording and you want to participate in the discussions, you can join us on our Discord server. Oh, yes. And then there's a link for the Discord. So aka.ms4 slash itopstalk-discord. Great point, Pierre. Next up, events. Yes. This is a big one. Yes. So there's a Windows event that's happening next week. We can't say much about it other than you definitely do not want to miss this event as they're going to be talking about what's next for Windows. And I'm interested. I tell you, I've been a big fan of Windows for a long time and I've seen it evolve. First version of Windows had used, Pierre. What was that? Two points. No, one point something. Were you on the one? Yep. Because I remember three dot one. That's the furthest back I go. And then that and creating the MS-DOS boot disks to play Doom. Yeah, no, I was on the one of the first versions and I installed that at the time, my 286, which with the turbo button. You have the turbo button, yes. I had the turbo button. So my 286 was, and this was like, I was looking at it going, wow, what am I going to do with this? This is ridiculous. Yeah, we've come in a long way since then. You've come a long way. Now you have devices that you wear that have an OS that you would interact with, which is very similar to what you would do on your desktop, but different. So that's why this is a very important event to see what Microsoft has in store for Windows. I'm very intrigued by it. Also today is patch and switch. Actually not patch and switch, testing and production. Exactly. Right, testing and production. So yourself and audio guy, Steve, AKA Jared Shockley, will be on to talk about, I think you're talking about audio today, right? We are talking about multiple things. Work updates, updates in general for streaming setup in terms of OBS and others. And of course, we're also going to talk about audio as part of that update. Last but not least, Hello World has their last episode today before they go on summer break. And we've been asked to be on the show and to talk about ASID update on Hello World. So that should be interesting as well. Yeah. Sorry, I'll be in testing and production during that time. So you're going to be flying solo on that one. On Hello World, yes. Well, it's myself and whoever's hosting the Friday episode, which would be cool. So it'd be different. Yeah, it was on last Friday. It was actually pretty fun. Oh yeah, I've been having fun. The Hello World piece, I love the fact that the stories are so different. And it's a side of technology as opposed to focus on technology, which is really cool. Yeah, and it's our opportunity when we participate to bring a different perspective to that show. The perspective of the IT and OBS audience to a show that is typically more geared toward developers. So that's kind of, it brings a nice balance to it. Let's talk about our Microsoft Learn module of the week. So the tie into our beginning banter in regards to Sonya's post, if a VM is not running in your Azure tenant, is are you being charged? Here's a great hands-on module that you can go through specifically to VM. So understanding Windows Server IaaS virtual machine cost management. If you're looking to understand the management breaks that you would get, the cost breaks that you would get in terms of the different facets of running the VM inside of Azure and the different ways to shut it down through PowerShell CLI or through the OS itself. Here's a great way to upscale. I know you love that word. Upscale on the aspects of what you can do to manage your VM costs and your IaaS implementation on Azure. Yes, I do not like that term, but anyway. Yeah, it's a great way. It's an everybody who's starting or even people who've been using it for a while. It's a great refreshing or refresher on all of the things that you need to look at when you're looking at your cost management of VMs in Azure. Typically you end up with people deploying stuff in Azure and they don't completely understand it. And then at the end of the month, they get the bill and they go, oh, wait a minute, I wasn't really, there was no load on it at night. Well, they were still running. Producer Pierre, if they want to get a hold of you, what's the best way to get a hold of you? I have been and always shall be at Wired Canuck. And if you want to get a hold of me for some reason, you can get a hold of me on Twitter at Wired Us Live. Everybody have a great weekend and we'll see you all next week. Cheers.