Azure – Page 13 – Buchatech.com

Azure Stack Development Kit (ASDK) Deployment Step by Step

July 12, 2017 by sbuchanan

At Microsoft Inspire Microsoft announced the Azure Stack Development Kit (ASDK) as a replacement to the POC and the general availability of the production Azure Stack named Azure Stack Integrated Systems. The Azure Stack Development Kit is here to stay. This will remain single node and should be used for trying out Azure Stack. You can develop your ARM templates and or applications on it and they will work on a production Azure Stack. The Azure Stack Integrated Systems are the ones that you buy from the OEM partners HP, Lenovo, Dell and soon to be Cisco, Avanade, and Huawei.

The ASDK install has improved 1,000 times over the previous TP’s of Azure Stack. I am going to detail the steps in this blog post. The steps start after you have downloaded the Azure Stack cloudBuilder.vhdx. Here we go:

PREPARE AZURE STACK HOST SERVER:

First off download the Azure Stack tools onto your Azure Stack host server. Just download all the tools as you will need all of them at some point. They can be found here: https://github.com/Azure/AzureStack-Tools

I put these in a folder on the C: drive named ASTools. I extract them and place them in the root.

Open up an elevated PowerShell window, navigate to your Astools folder and run the asdk-installer.ps1 script. Next a GUI wizard will pop-up.

Click on Prepare Environment.

Point it to your cloudBuilder.vhdx and click Next.

Put in the host servers local admin password. Make sure this matches the Azure account you plan to use.

Select the other options as you see fit.

It will run for a while creating the unattended file for Windows Server 2016.

Once it is done click Reboot now.

DEPLOY AZURE STACK DEVELOPMENT KIT:

Next lets deploy Azure Stack. After the server has rebooted log onto your AS server. Use the localhost\administrator account and the password you set.

Once again from PowerShell run asdk-installer.ps1. A GUI wizard will come up. Click on Install.

Select Azure Cloud (Azure Active Directory) or ADFS. Put in your directory and password.

Verify and select the correct NIC.

Select DHCP or put in your static IP settings.

It will verify the network settings.

You will see the PowerShell deployment script that will be run. Click on Deploy!

The PowerShell deployment will kick off in a PowerShell window.

After a little bit (1-2 minutes) an Azure login window will ask for your Azure account creds. This is the account ASDK will be deployed under.

NOTE: We still have the log folder and files under CloudDeployment on the C drive.

A few hours later and there it is successfully!

Having been involved with Azure Stack since TP1 and losing about a week to deploying Azure Stack TP1 this is a much….much better deployment experience. Nice work Microsoft Azure Stack team!!!

OMS and Cherwell ITSM Integration

May 25, 2017 by sbuchanan

Microsoft recently released public preview of OMS and ITSM integration. Here is the link for that announcement:

https://blogs.technet.microsoft.com/msoms/2017/05/11/it-service-management-connector-public-preview

Microsoft has built an ITSM connector in OMS. This new ITSM connector can connect to many ITSM solutions out there. The ITSM solutions it can connect to are:

System Center Service Manager (SCSM)
Cherwell
ServiceNow
Provance

This new ITSM connector is bi-directional meaning work items can flow from the ITSM solution into OMS and OMS can create work items in the ITSM solution such as incidents, alerts, and events. Hopefully in the future OMS could be used to populate a CMDB and even create application maps from OMS’s Service Map.

I wanted to give this a test run with a test Cherwell instance that I have. There is official documentation for the integrations. The documentation is good however after setting this up I did find that there could be a few more steps spelled out as well as screenshots with the Cherwell piece.

Needed settings from Cherwell:

Before you set the connection in OMS go and get the information you will need. So you will need a username and password of an account that has access to Cherwell, the Cherwell URL, and a Cherwell Client ID.

If you don’t know your Cherwell URL you can get this from the Cherwell client console. Launch Cherwell.

Before you login you can edit the connection to see the URL as shown in the screenshot. You will want to copy this to use in the OMS ITSM connector setup.

Note that you do not want to copy the entire URL. Only copy to the .com like https://test.demo.cherwell.com.

Next we need to generate the Client ID. The Client ID is basically a generated string called the client key used for connecting to Cherwell’s API. To get this client ID Launch the Cherwell Administrator console.

A window will pop up and you will need to click on the green plus to create a new one. Give it a name and copy out the Client Key.

Copy this as you will need it later.

Setup in OMS:

Next log into OMS and add the ITSM Connector preview. It is shown in the screenshot below.

After this has been added go to your OMS settings screen click on Connected Sources>ITSM Connector and then click on Add New Connection.

Select Cherwell for the connection type add in your Cherwell settings and save it. If everything worked your OMS is now connected to Cherwell.

Exploring the ITSM Connector:

Next let’s check things out in OMS. Before I did that I first went and created a new incident so I could see this flow over into OMS. So I created the following over in Cherwell:

After doing that I went back into OMS and kicked off a sync because I did not want to wait.

The connector picked up my new incident right away. You can see the dashboard ITSM tile has 2 incidents.

After clicking into this I am brought to the full ITSM dashboard. I then clicked on the Incident tile.

I was then brought to the incident dashboard and I could see the new incident I created.

I clicked on the new incident and it brought me to the OMS search with the details of the incident. Very cool!

I am excited to see cool stuff like this in OMS and integration with many ITSM tools. Look out for more blog posts in the future about ITSM Integration in OMS as well as in Azure Stack.

Speaking at OSCON and MMS May 2017

May 7, 2017 by sbuchanan

May is a busy month for me with the opportunity to speak at both OSCON – Open Source Convention (https://conferences.oreilly.com/oscon/oscon-tx) and MMS – Midwest Management Summit (https://mmsmoa.com)! OSCON is May 8th through the 11th in Austin Texas. MMS is May 15th through the 18th in Minnesota.

At OSCON I will be presenting on “How to Motivate Technical Employees” with friend and fellow Microsoft MVP Samuel Erskine – @samerskine. This will be on Thursday, May 11, 2017 at 4:15pm–4:55pm. This session is for CIOs, CTOs, IT directors, and IT managers and will cover how to retain your top talent and give you five ways to motivate technical employees. Come to this session to learn the secret sauce for keeping employee’s engaged! Here is a link to the session: https://conferences.oreilly.com/oscon/oscon-tx/public/schedule/detail/57374

At MMS I will be presenting three sessions! These sessions are:

“Awesomize your Azure Stack Deployments with Azure Stack Tools” with fellow MVP Mikael Nystrom -@mikael_nystrom.

Link: https://mms2017.sched.com/event/AUae/awesomize-your-azure-stack-deployments-with-azure-stack-tools

“Azure Operationalized” with fellow MVP Natascia Heil – @NatasciaHeil.

Link: https://mms2017.sched.com/event/AUbn/azure-operationalized

“Backup is Dead! Restore is Born in the Cloud!” with fellow MVP Robert Hedblom -@RobertandDPM.

Link: https://mms2017.sched.com/event/AUaR/backup-is-dead-restore-is-born-in-the-cloud

These conferences will be lots of great community fun! Hope to see you there.

Sys Admin to Cloud Admin…ITSM to CloudOps…On-Prem to Azure Stack/Azure

April 7, 2017 by sbuchanan

A while back I posted a blog titled “Surviving the future of IT as an IT pro”. In that blog post I set out to share my opinion on where IT is headed and what you should focus on as an IT pro going forward. I guess this post could be considered part 2 however in this post I will focus more on where things are heading as a whole.

So what is this blog really about? It is about “the Transition from ITOPS & ITSM to CloudOps via Azure Stack (Hybrid Cloud) powering DevOps and becoming core to the Digital Transformation of business” that is happening. Whew…..Ok, a lot was said in that previous sentence. J Let’s break it down.

Transition from ITOPS & ITSM to CloudOps

There has been this transition in IT for a while to increase the density in data centers. This was started with the wide adoption of the hypervisor (VMWare, Hyper-V, Citrix Xen etc…). The goal is to get more out of existing and less physical hardware. Think about 1 physical server hosting hundreds of virtual servers. Things have since accelerated at a fast pace. We now have containers, PaaS, and serverless. With these newer technologies, the density is even greater.

The real power behind cloud is software defined everything. With software, defined environments AKA cloud a new skillet and a different way of thinking about managing operations is needed. This new skillset and new way of thinking for the operationalization of cloud is known as CloudOps. IT Operations and IT Service Management do not go away with CloudOps. The evolution of ITOPS and ITSM become CloudOps. The best parts of ITOPS and ITSM (ITIL) funnel into CloudOps used for operating clouds.

Hybrid Cloud (Azure Stack)

Hybrid Cloud is going to be a huge part of cloud initiatives in many organizations for years to come. You can see this on the Gartner reports(http://www.gartner.com/newsroom/id/3354117), Right Scale reports (http://www.rightscale.com/blog/cloud-industry-insights/cloud-computing-trends-2017-state-cloud-survey) and based on the investments the major cloud players are making to build the best Hybrid Cloud solutions.

Hybrid Cloud Is the Preferred Enterprise Strategy, but Private Cloud Adoption Fell

From Rightscale “Cloud Computing Trends: 2017 State of the Cloud Survey” Report:

http://www.rightscale.com/blog/cloud-industry-insights/cloud-computing-trends-2017-state-cloud-survey#hybrid-cloud

Recently IBM and Red Hat announced their launch into the Hybrid Cloud space.

(http://www.networkworld.com/article/3182989/cloud-computing/ibm-red-hat-an-open-source-hybrid-cloud.html)

A while back Amazon and VMWare announced their launch into the Hybrid Cloud space.

(http://www.businesswire.com/news/home/20161013006574/en/VMware-AWS-Announce-Hybrid-Cloud-Service-%E2%80%9CVMware>)

Microsoft was the first to jump into the Hybrid Cloud space and is the only company that has a 100% true Hybrid Cloud solution. Both VMWare/Amazon and IBM/Red Hat have solutions that run private cloud on public cloud. The private cloud solutions are being retrofitted to run in public cloud as the framework for their Hybrid Cloud solutions. These are not consistent cloud platforms running the same exact bits on bare metal on-premises and in the cloud like Microsoft’s Azure Stack solution. Azure Stack is the same bits in the public cloud and on-premises down to the bare metal.

IBM and Amazon jumping into the Hybrid Cloud space is more proof this will be a large area of growth in IT for years to come. I wonder if Google will decide to jump into the Hybrid Cloud space at some point and what their strategy will be.

DevOps powered by Azure Stack and CloudOps

Azure Stack serves as a catalyst to help move DevOps initiatives forward within organizations. With Azure Stack’s comes the native ability to run the environment using Infrastructure as code, continuous integration, continuous delivery, microservices, integration with source control systems, and more. All of the aforementioned are a part of DevOps.

Along with Azure Stack is the need to run the environment using a CloudOps model. Here is a list of concepts that drive CloudOps:

Extreme Hardware Standardization
Software Defined Everything
Extreme Automation
Focus on Zero Downtime
Self Service
Measured Service
Multitenancy

CloudOps is overall focused on business applications critical for running the business through the continuous operations of clouds. CloudOps leaves business unit projects to DevOps. CloudOps instead focuses on the delivery of the the cloud infrastructure to support self-service leveraged by DevOps teams.

David Armour of Microsoft often shares great information on CloudOps and what it means. You can follow him on twitter here: https://twitter.com/Darmour_MSFT

CloudOps supports DevOps and DevOps is core to Digital Transformation

Digital Transformation is the accelerating transformation of the way businesses do business from traditional ways often brick and mortar to the digital front through the use of digital technologies. Businesses are shifting to meet their customers and employees where they are today on digital platforms. In the business world, today it is well known that you must innovate and grow through the use of technology or become obsolete and left in the wake of disruptive companies that are leveraging technology to meet their customers on the digital front.

Examples of digitally transformed company’s vs non-digitally transformed companies are:

Netflix vs Blockbuster
Amazon vs Target, Best Buy, Macy’s
Airbnb vs Wyndom hotels
Uber vs Taxi Companies

Digital Transformation is critical to business and IT departments need to be a core driver to help organizations move forward on the digital transformation front. Digital Transformation is the new Industrial Revolution of business today with CloudOps/DevOps being the Assembly line that will bring innovation to the business.

Through DevOps businesses can bring digital services to the market at very fast rates and can pivot quickly as needed to beat and stay ahead of the competition meeting the customers’ demands in an agile way. CloudOps allows the scale and another point to pivot on at any time to redirect in a new direction as needed by the business in an agile manor.

Through a Hybrid Cloud solution like Azure Stack things IoT, Microservices, extreme automation, hyper-scale, and agility can be realized for the business empowering Digital Transformation from the core.

The transition of the IT Pro to Cloud Pro

Ok. That was a lot of information and background on CloudOps, DevOps, Digital Transformation and Hybrid Cloud. You may be asking yourself at this point where does the IT Pro fit into the picture? Let me answer that for you and take you on a tour of Azure Stack to prove why as an IT Pro you should start working with it today!

The path for an IT Professional when moving from traditional IT into a Hybrid Cloud world consists of:

	A cloud administrator can configure and manage resource providers, tenant offers, plans, services, quotas, and pricing.
	A tenant purchases (or acquires) services that the service administrator offers. Tenants can provision, monitor, and manage services that they have subscribed to, such as Web Apps, Storage, and Virtual Machines.

Those cloud roles fit in a new world of CloudOps including Cloud architect, engineer, and administrator. Being a part of CloudOps requires a different mindset. Think about dynamic shifts such as software defined everything and extreme standardization. More concepts and technologies that a cloud role requires an understanding of are:

IaaS
PaaS
Software Defined Data Center technologies
Automation
Source Control Systems
Business Intelligence (Showback/Chargeback)
High Availability technologies
Backup and Disaster Recovery
Scaling technologies
Containerization
Server less technologies
Cloud Security
Both Linux and Windows
Self-Service (Service Catalog)
Multitenancy technologies
Tenant administration
And more

Ok. Now let’s jump into some example of CloudOps tooling in Azure Stack. First off, we as a cloud admin you need to know how to perform management of tenants (customers). Here is an example of a dashboard for doing this in Azure Stack:

In Azure Stack, you will need to know and understand the administration of managing the cloud itself. This includes many things some of them being management of a region/s, resource providers that contain the services you can offer up to tenants, along with monitoring, high availability, and backup of the cloud. Below is an example of administration in Azure Stack at the cloud model layer of CloudOps.

We already mentioned monitoring. There is monitoring of the cloud environment itself but there also is a need to monitor the resources being consumed by the tenants. One of the great things about Azure and Azure Stack is the out of the box monitoring and health diagnostics of IaaS virtual machines. I am a SCOM guy and have done a lot of SCOM projects. SCOM works well and serves a purpose but the out of the box monitoring in Azure and Azure Stack is amazing in the ease of turning it on. Once turned on it just works and has very nice visuals to see and work with as shown in the following screenshot. As a cloud administrator, you need technology to be easy so that you can move away from complex setups and troubleshooting the monitoring solution and move to monitoring the resources.

One of the best benefits about Hybrid Cloud is the consistency between public and on-premises cloud. In the following screenshot news updates on Azure and Azure stack are the same. 🙂 Another huge point of consistency between Azure and Azure Stack is the ability to view, deploy and run items from the Azure marketplace in Azure Stack. This is called marketplace syndication.

Azure

Azure Stack

Azure Stack is set to release in 2017. I want to highlight some of the services already in Azure Stack and more coming to Azure Stack that can be offered in your Service Catalog to tenants.

Already in Azure Stack as of TP3:

SQL PaaS
MySQL PaaS
Web Apps PaaS
Computer IaaS
Virtual Machines (Linux or Windows)
VM Scale Sets
Storage
Networking
PaaS: Storage
Key Vault
Management of Azure Pack virtual machines
Marketplace Syndication

Coming to Azure Stack at some point:

Microservices
Service Fabric
Cloud Foundry
Blockchain
Container Service
IoT

Another big part of CloudOps is being able to measured services that are being consumed. Measured Service can translate to show back or charge back. Measured Service is the ability to track the usage of resources down to the individual resource level. With Azure and Azure Stack resource management (ARM) model resources are carved out and placed into resource groups. In ARM, each resource has an associated cost that is tracked via the usage. There is full role based access around resources and resource groups. Resources and resource groups can be tagged and each resource or resource group’s usage can be tracked and displayed on business intelligence reporting or a dashboard like shown in the following screenshot.

That concludes this blog post. I hope I was able to shed some light on the transition from IT Pro to cloud pro, from IT Ops/ITSM to CloudOps and showcase the power of Hybrid Cloud via Azure Stack. Stay tuned for more exciting stuff coming from Azure Stack.

Azure Stack POC Hardware

March 5, 2017 by sbuchanan

I have been asked several times what I use for my Azure Stack rig and where I got the hardware from. I am going to share in this post what I use to run my single node Azure Stack POC. I bought all parts from newegg.com. Here is a list of the parts:

Motherboard: MSI X99A SLI KRAIT EDITION LGA 2011-v3 Intel X99 SATA 6Gb/s USB 3.1 USB 3.0 ATX Intel
Video Card: EVGA GeForce 210 DirectX 10.1 512-P3-1310-LR 512MB 32-Bit DDR3 PCI Express 2.0 x16 HDCP Ready Low Profile
Power Supply: EVGA 750 BQ 110-BQ-0750-V1 80+ BRONZE 750W Semi Modular Includes Power On Self Tester
Processor: Intel Core i7-5820K Haswell-E 6-Core 3.3 GHz LGA 2011-v3 140W BX80648I75820K Desktop
NOTE: I was not paying attention when I bought this. Azure Stack needs 12 cores. I am able to work around this and have not run into problems yet. When I get a chance and $$$ I will upgrade this.
3 SSD Hard Drives: PNY CS1311 2.5″ 960GB SATA-III (6 Gb/s) TLC Internal Solid State Drive (SSD) SSD7CS1311-960-RB
NOTE: I bought a couple of more Kingston brand SSD’s. I use these for the OS and general storage.
Memory: G.SKILL Ripjaws 4 Series 128GB (8 x 16GB) 288-Pin DDR4 SDRAM DDR4 2800 (PC4 22400) Intel X99 Platform Extreme Performance Memory Model F4-2800C15Q2-128GRKD
Case: Corsair Carbide Series Air 540 CC-9011034-WLED Silver Steel ATX Cube Computer Case

As you can see this is generic hardware. The cost of this hardware was just over $2k USD. I have been running Azure Stack since TP1 on this hardware and I am currently running TP3. This is a personal lab for just me and Azure Stack runs well on my hardware. Don’t let a lack of hardware stop you from diving into Azure Stack. As you can see from this post it does not take much to pick up some parts and get going.

I do also run another Azure Stack POC on much better hardware at work. I can’t wait to get a multi-node environment on one of the hardware providers (Cisco, Dell, Lenovo, or HP) platform.

Here is what my rig looks like complete with Azure Stack and other stickers :-).

Before it was built:

After it was built and running:

Backup Strategy should include Security

March 2, 2017 by sbuchanan

Planning for protection as a part of an IT Service Continuity plan often takes into consideration backup of applications and data as well as restore. But what about security?

When planning for protection of applications and data in your environment security should right up there in the forefront. “Backup Security” should be a key part of the plan.

Security in the context of backup can be thought of #1 as securing the backups, and #2 backups being used as an added measure for security breach mitigation. Let me break this down further.

In regards to securing backups you want to do things like encrypt backup data as it travels offsite, encrypting backup data at rest, being able to protect encrypted data, requiring security pins or further authentication of admins and more.

In regards to backup as an added measure for security backup becomes a direct part of Security planning in organizations. Sometimes when security measures fail backups are the only thing that can save you as a last resort. Backups are commonly becoming a way to recover from ransomware attacks as an alternative to paying the hackers. Here is a real world example.

Recently an unnamed hosting providers entire data center became hostage to a ransomware attack. This hacker got in due to a mistake of one of the system admins (more on how to protect at this level later) and basically had full domain admin rights to everything. Keep in mind majority of the servers in this scenario are for customers.

In this case the hosting provider had two choices. Option #1 go to the dark web via a tor network and pay a ton of money in bitcoin for the decryption key. Option #2 Restore everything from offsite backups and pray.

This hosting provider went for option #2 and thank goodness it worked. In this case if it weren’t for a solid offsite backup solution this hosting provider would have been up a creek without a paddle.

It is becoming more common that ransomeware will actually target backups because these are a high target and hackers understand this is a last resort for companies to save themselves. If the backups are deleted there is no other choice but to pay the ransom. This raises the security level of the backups. Administrative actions on backups need an extra layer of security.

Microsoft Business Continuity products help with not only protection but also security. These products consist of System Centers Data Protection Manager (DPM) and Operations Management Suites Azure Backup (AB) and Azure Site Recovery (ASR). In this post I am only going to touch on DPM and AB.

Some exciting things have been happening with Azure Backup and Data Protection Manager to ensure security is front and center as a part of your enterprise backup solution. Microsoft’s goal with the backup security is to provide prevention, alerting, and recovery.

More about this including a video can be found here:
https://azure.microsoft.com/en-us/blog/azure-backup-security-feature

Just yesterday DPM update rollup 12 for 2012 and update rollup 2 for 2016 was announced. Along with UR2 comes some enhanced security features for DPM. These will be called out later in this blog post. Microsoft has rolled out some great security features to both across hybrid clouds. I will go ahead and break these down.

– Azure Backup –

Encrypted backup data at rest
Described in DPM section.

Security PIN
With Azure Backup you can require a security pin for sensitive operations such as removing protection, deleting data, or changing other settings in Azure Backup itself such as changing a Passphrase.

Azure Backup also has some other security measures in place like a minimum retention range to ensure a certain amount of backup data is always available and notifications upon critical operations to subscription admins or others as specified.

NOTE: These security features are now also available in DPM with the UR’s (UR 12 for 2012 and UR2 for 2016) announced yesterday. When an administrator changes the passphrase, or delete backup data, you need to enter the PIN if you have Enhanced Security Enabled. Also, there is a minimum retention range of 14 days for cloud protected data that is deleted.

MFA
MFA is Multi-Factor Authentication. Microsoft has MFA available as a part of Azure Active Directory. Within Azure Backup you can configure it to require MFA of admins when performing critical operations. By enabling MFA you would then ensure via authentication from a second device usually physical to the user that they are who they say they are.

NOTE: When you enable security settings they cannot be disabled.

Ransomware attacks
Described in DPM section.

– Data Protection Manager –

Backup data encrypted during offsite transfer
When data is sent from DPM to Azure Backup it is encrypted before it even leaves your four walls. Data is encrypted on the on-premises server/client/SCDPM machine using AES256 and the data is sent over a secure HTTPS link.

Encrypted backup data at rest
Once backup data is on Azure it is encrypted at rest. Microsoft does not decrypt the backup data at any point. The customer is the only one with the encryption key that can decrypt the backup data. If this key is lost not even Microsoft can decrypt your backup data. This is very secure.

Protection and recovery of encrypted computers
The release of Hyper-V on Windows Server 2016 included a new feature known as Shielded virtual machines (VM’s). This feature essentially utilizes Virtual Trusted Platform Module (vTPM) technology and BitLocker to encrypt a VM to encrypt virtual machines at the virtual layer. This means if a VM is physically copied off a Hyper-V host whoever has the VM will not be able to get to the data on the virtual hard drive.

With the release of DPM 2016 it supports protecting Shielded VM’s. DPM can protect Shielded VM’s regardless if they are VHD or VHDX. This is great news because as a secure organization you should want to encrypt your virtual machines and DPM can protect them. This gives you an added layer of security on top of having backups.

Ransomware attacks
In today’s world ransomware attacks are a common thing. These type of attacks are targeted at small, medium, and large enterprise businesses. No company is too small or too big to be put in the crosshairs of ransomware attacks. A well-known attack is Cryptolocker.

As mentioned before in this blog post backups are an alternative to paying the ransom of a ransomware attack. They key here is to ensure you have a solid offsite backup in place such as Azure Backup. Having that offsite backup will ensure you can get your data back even if the ransomware attack get ahold of your onsite backup data.

I even go as far as to recommend sticking to the 3-2-1 rule (3 copies of backup data 2 offsite and 1 onsite). This way if something happens to one of your offsite copies of data you have another one. It may seem overkill to have 2 offsite copies but you would be surprised how often offsite backup data is accidently destroyed.

So there you have it. Security is a critical part of any backup solution. It is clear that Microsoft realizes this based on the security enhancements they have made to both Azure Backup and Data Protection Manager 2016. Their goal is to ensure both backup solutions are enterprise ready. I have been working with DPM for years and Azure Backup as soon as it came out. I know the team behind these products have a lot of new features and functionality planned for the future of these products and I am looking forward to it.

Monitor Azure Stack Fabric with OMS

January 2, 2017 by sbuchanan

I wanted to monitor my Azure Stack environment with OMS. This would include only the Azure Stack fabric servers and the host. I did not want to manually install the OMS agent on all of these servers especially since the Azure Stack fabric is a set of known servers. So I decided to put together a quick PowerShell script to handle the install of the OMS agents including the workspace ID and key. Here are details for the script:

.SYNOPSIS
This script can be used to install OMS agents on all of the Azure Stack Fabric servers. This has been tested with TP2.

.DESCRIPTION
This script can be used to install OMS agents on all of the Azure Stack Fabric servers. This has been tested with TP2. This script can be run from PowerShell ISE or a PowerShell console. It is recommended to run this from an elevated window. This script should be run from the Azure Stack host. Ensure you are logged onto the Azure Stack host as azurestack\azurestackadmin. This script allows you to input your OMS workspace ID and key. The Azure Stack Fabric servers that this script will attempt to install on is:

“MAS-Con01”,

“MAS-WAS01”,

“MAS-Xrp01”,

“MAS-SUS01”,

“MAS-ACS01”,

“MAS-CA01”,

“MAS-ADFS01”,

“MAS-ASql01”,

“MAS-Gwy01”,

“MAS-SLB01”,

“MAS-NC01”,

“MAS-BGPNAT01”

Fabric servers can be added or removed from the array list if desired. The script will look for the OMS agent (MMASetup-AMD64.exe) in C:\OMS\ on the Azure Stack host. Ensure you create an OMS folder on your Azure Stack host and download the OMS agent to it. This script also copies the OMS agent to C:\Windows\Temp on each Fabric server. Ensure there is enough free space on the C drive on all of your fabric servers.

.PARAMETER OMSWorkSpaceID
This is Guid ID for your OMS workspace, it can be found in the OMS portal at: https://mms.microsoft.com >> Overview >> Settings >> Connected Sources >> Windows Servers

.PARAMETER OMSKey
This is the OMS API key for your OMS workspace. You can use the primary or secondary key. These keys can be found in the OMS portal at:
https://mms.microsoft.com >> Overview >> Settings >> Connected Sources >> Windows Servers

.INPUTS
None

.OUTPUTS
None

.NOTES
Script Name: AzureStackFabrickOMSAgentInstall.ps1
Version: 1.0
Author: Cloud and Data Center Management MVP – Steve Buchanan
Website: www.buchatech.com
Creation Date: 1-1-2017
Purpose/Change: Install OMS agents on Azure Stack Fabric servers.
Updates: None

.EXAMPLE
.\AzureStackFabricOMSAgentInstall.ps1 -OMSWorkSpaceID “20d4dd92-53cf-41ff-99b0-7acb6c84beedsr” -OMSKey “aazedscsjwh52834u510350423tjjwgogh9w34thg2ui==”
#>

The script can be downloaded here:
https://gallery.technet.microsoft.com/Azure-Stack-Fabric-OMS-3dac666c

To kick off the script run from PowerShell ISE or a PowerShell console. If you run from ISE you will be prompted for the workspace ID and the key. If you run from a PowerShell console run this syntax to kick it off:

.\AzureStackFabricOMSAgentInstall.ps1 -OMSWorkSpaceID “YOURWORKSPACEID” -OMSKey “YOUROMSKEY”

The script will kick off, building an array of the Azure Stack VM’s, looping through each of them to copy over the OMS agent, and then install the OMS agent setting the OMS workspace ID and key.

The script will detect if an OMS agent is already installed and will skip that server as shown in the following screenshot.

Otherwise the script will install the OMS agent as shown in the following screenshot.

The following screenshot shows the script running in a PowerShell console vs ISE.

You will be prompted when running the script for credentials. Use Azurestack\azurestackadmin as shown in the following screenshot.

After the OMS agent is installed you should be able to log onto any of the Azure Stack VM’s and see the OMS agent in control panel as shown in the following screenshots.

You can also log onto OMS and see your Azure Stack servers listed under connected computers.

Azure Stack fabric servers wire data:

My Azure Stack host in OMS Service Map:

Happy Stacking and OMS’ing!

Azure or Azure Stack “Write Once, Deploy Anywhere” Update

December 20, 2016 by sbuchanan

A while back I wrote a blog post about being able to take one IaaS VM Azure Resource Manager (ARM) template and deploy it to both Azure or Azure Stack. This blog post included a JSON file and the PowerShell to do this. The idea for that came from needing to set up a cool and working demo for MMS 2016 and the need to showcase the power of Microsoft’s HybridCloud. Here is a link to that original blog post:

Write once, deploy anywhere (Azure or Azure Stack)

Today I have finished updating the PowerShell and ARM template/JSON file to be more streamlined and to work with TP2. Here is the link to download these:Here are the updates:

https://gallery.technet.microsoft.com/Create-VM-on-Azure-or-3c6d0420

Here are the updates:

The JSON and PowerShell script have been modified to work with Azure Stack TP2.
This script now utilizes the connection PowerShell module AzureStack.Connect.psm1 from Azure Stack tools.
This is included with the download of this script and JSON file on TechNet Gallery.
The script is hard coded to look locally to import the AzureStack.Connect.psm1 module.
Streamlined the JSON file and PowerShell script.
The script no longer prompts for the publicDNS name. It is now automatically set to the same as the vmname.
The script no longer prompts for the storage account name. It is automatically set to vmnamestorage.
The script no longer prompts for the resourcegroup name. This is now automatically set to vmname-RG.
By default this script now uses a JSON file hosted on Github. This is set in the $templateFilePath variable as shown on the next line.
To keep it to the local directory just use the JSON file name.

GITHUB: $templateFilePath = “https://raw.githubusercontent.com/Buchatech/Azure-AzureStackVM/master/AzureandAzureStack.json”
LOCAL: $templateFilePath = “AzureandAzureStack.json“

This will be my last blog post of 2016. See you next year folks…..

Happy Stacking!

Resource Group Clean-up in Azure Stack

December 12, 2016 by sbuchanan

If you are like me, you end up creating a ton of resource groups in Azure Stack when testing things out. I needed a way to delete them without having to click one each one via the portal. The best option of course is to leverage PowerShell. I threw together some PowerShell to handle this. I came up with two options #1 can be used to delete a bunch of RG’s that have a common name. For example, I had a bunch of VM00* resource groups. I use the script to go loop through and delete all resource groups with VMO in the name. Option #2 pop’s up a GUI window so I could select the RG’s I wanted to delete. It put them in an array and then looped through to delete them in one shot.

This is great because I can kick this off and go do something else. I will share both below in this blog post along with some screenshots. I won’t have a download for the PowerShell syntax so just copy from this post if you want to use it. Be sure to use AzureStack.Connect.psm1 for connecting to your Azure Stack environment before running any of the following code.

Code:
#1

#Create Variable of RG’s with common name
$Resourcegroups = Get-AzureRmResourceGroup | where {$_.ResourceGroupName -like (‘*VM0*’)}

#Create array of RG’s
$RGLIST = $Resourcegroups.ResourceGroupName

#Loop to remove each resource group in the array
ForEach(
$rg in $RGLIST
)
{
Get-AzureRmResourceGroup -Name $rg -ErrorAction SilentlyContinue | Remove-AzureRmResourceGroup -Force -Verbose
}

This image shows the array of RG’s that will be looped through. I highlighted vm003rg in the array and in the PowerShell status message.

The following screenshot shows VM003RG being deleted in the Azure Stack portal.

#Create Variable of RG’s from GUI selection
$selectedrgs = (Get-AzureRmResourceGroup | Out-GridView ` -Title “Select ResouceGroups you want to remove.”` -PassThru).ResourceGroupName

#Loop to remove each resource group in the array
ForEach(
$rg in $selectedrgs
)
{
Get-AzureRmResourceGroup -Name $rg -ErrorAction SilentlyContinue | Remove-AzureRmResourceGroup -Force -Verbose
}

After running the Create Variable of RG’s from GUI selection part of the code a window as shown in the following screenshot will pop up. Select the RG’s you want to remove, click Ok and they will be placed into an array.

Below if the output of the array. Run the Loop to remove each resource group in the array part of the code and each of the RG’s will be removed.

I have also used this when a resource group would not delete from the portal. On some stubborn resource groups I have had to run this a couple of times. This is a short post. I hope this helps someone out!

OMS: Service Map overview

November 28, 2016 by sbuchanan

Recently the Operations Management Suite (OMS) team at Microsoft announced the private preview of Service Map in OMS formally known as Application Dependency Map. Service Map has been a long awaited feature in OMS. Service Map is a feature that is a part of OMS that discovers and maps Windows & Linux app and system dependencies. Service Map displays these dependencies in application maps within OMS. Service Map did not start with OMS. It actually started as a standalone product named Fact Finder and later was integrated with SCOM. The integration of FactFinder with SCOM allowed Bluestripe to automatically create Distributed Applications in SCOM. Well Microsoft acquired BlueStripe and the rest is history.

In this post I will set out to explore and break down Service Map, how it is installed, info about the agent, how it works, key points about it, how the data flows and more. NOTE: Click on any of the images in this post to display larger in a new window. Also this post is my first effort in taking one of my PowerPoint’s and converting into a post! The following graphic describes some of the benefits of having application maps including in your monitoring solutions along with information about FactFinder:

Now let’s take a look at what Service Map does and how it looks.

Now let’s take a look at one of the Service Maps aka Application Maps in OMS. Notice on the left hand side the breakdown of the interface. In Service Map there is a focus machine in the center. There are front end and back end connections into that focus machine. These are the dependencies flowing in and out of the focus machine giving the mappings. Notice on the left-hand side you can control the time controls and select either a Windows or Linux machine from the list. Finally, on the left-hand side are the details of the current selection. The current selection can be a machine or process.

Also notice that SM integrates with Change Tracking, Alerts, Performance, Security, and updates. What this means is that when you have a focus machine selected you can click on the corresponding solution on the right hand. When you click on the solution i.e. updates or security the update or security dashboard widget will be shown and you can drill down from there for further detail.

A common question that comes up when discussion Service Map is how does it work. The following graphic displays the process from the solution add to the actual mapping within OMS.

Other key information about Service Map is detailed in the following graphics.

The next graphic looks at deploying the SM agent and locations for logs. The process is as simple as downloading and installing the agent from OMS.

Here is some more critical information you need to know about the SM agent.

This next graphic details how Service Map dependency data flows into OMS.

At this current time Service Map supported Operating Systems at this time are:

Windows	Linux
Windows 10 Windows 8.1 Windows 8 Windows 7 Windows Server 2016 Windows Server 2012 R2 Windows Server 2012 Windows Server 2008 R2 SP1	Oracle Enterprise Linux 5.8-5.11, 6.0-6.7, 7.0-7.1 Red Hat Enterprise Linux 5.8-5.11, 6.0-6.7, 7.0-7.2 CentOS Linux (Centos Plus kernel is not supported) SUSE Linux Enterprise Server 10SP4, 11-11SP4

Service Map’s computer and process inventory data is available for search in OMS Log Analytics. This is very cool as the log analytics and searching capability in OMS is powerful and most important very FAST. Having application components, service dependencies, and supporting infrastructure configuration data at your fingertips through the log analytics gives you a powerful troubleshooting and forensics tool. I am sure over time the query capabilities will be expanded to include even more.


Type=ServiceMapComputer_CL	Type=ServiceMapProcess_CL

A few Service Map Log Analytic query examples:

List the physical memory capacity of all managed computers:

Type=ServiceMapComputer_CL | select TotalPhysicalMemory_d, ComputerName_s | Dedup ComputerName_s

List computer name, DNS, IP, and OS version:

Type=ServiceMapComputer_CL | select ComputerName_s, OperatingSystemVersion_s, DnsNames_s, IPv4s_s | dedup ComputerName_s

List Process Map by process name:

Type=ServiceMapProcess_CL (ProductName_s=TeamViewer)

Thanks for reading and I hope you enjoyed this post on OM’s Service Map. Now go out and add the public preview right away.