Windows Server 2019 Module 6 Section 1: Planning for Failover Clustering Implementation

What is a failover cluster

a group of independent computer that work together to increase the availability and scalability of clustered roles.

Clustered servers are called

Nodes

How are clustered servers connected

physical hardware and software

What happens if one or more nodes fail

others take over the applications and roles the failed ones were providing

The process of nodes taking over for failed one is known as

failover

Failover clusters provide another service known as

Cluster Shared Volume (CSV)

What does Cluster Shared Volume (CSV) provide

provide support for a read cache, which can significantly improve performance in certain scenarios. Additionally, this File System can perform chkdsk without affecting applications with open handles on the file system.

What are some applications for failover clusters

High Availability applications such as Microsoft SQL Server and Hyper-V virtual machines (VMs)

High available clustered roles

Using Windows Admin Center provides the big advantage of

Having one interface for everything

What is an Active Node

a cluster currently running on it. A resource or resource group can only be active on one node at a time.

What is Cluster Resource

a hardware or software component in the cluster such as a disk, virtual name, or IP address.

What are Cluster Sets

Used to scale out topology by using the cloud. These enable virtual machine (VM) fluidity across member clusters within a cluster set and a unified storage namespace across the set.

What is a Passive Node

A node that doesn’t have a cluster running on it

Each Node needs how many network adapters

2

Each Node needs two network adapters for

One for Public networks and One for Private networks

The Public Network Adapter is connected to the

Lan or Wan for external or internal use

The Private Network Adapter is used for

communication between nodes

What is the communication between nodes called

heartbeat

What is a Resource Group

a single unit within a cluster that contains cluster resources. This is also called an application and service group

What is a Virtual Server

This consists of the network name and IP address to which clients are connected. A client can connect to this, which is hosted in the cluster environment, without knowing the details of the server nodes.

What is a Azure Cloud Witness

used to create a share quorum

What is a Cluster Quorum

This maintains the definitive cluster configuration data and the current state of each node. It also maintains each service, application group, and resource network in the cluster.

What is a Heartbeat

a health check mechanism of the cluster, where a single User Datagram Protocol (UDP) packet is sent to all nodes in the cluster through a private network to check whether all nodes in the cluster are online. One heartbeat is sent every second. By default, the cluster service will wait for five seconds before the cluster node is considered unreachable.

What are local disk referred to as

private storage

What is a Shared Disk

Each server must be attached to external storage. In a clustered environment, data is stored in this disk that's accessible by only the nodes in the system.

What is Storage Replica do

provides disaster recovery by enabling block-level, storage-agnostic, synchronous replication between servers. Using this in a wide range of architectures, including stretch clusters, cluster-to-cluster, and server-to-server.

What does a whiteness Disk or File Share do

used to store the cluster configuration information. They help to determine the state of a cluster when some or all of the cluster nodes can't be contacted.

What is Cluster Performance History

a new feature in Windows Server 2019 that gives Storage Spaces Direct administrators easy access to historical compute, memory, network, and storage measurements across an organization's servers, drives, volumes, VMs, and many other resources. The performance history is automatically collected and stored on a cluster for up to a year. The metrics are aggregated for all the servers in the cluster, and they can be examined by using the Windows PowerShell Get-ClusterPerf alias, which calls the Get-ClusterPerformanceHistory cmdlet, or by using Windows Admin Center.

What is Cross-domain cluster migration

migrates clusters from one domain to another by using a series of Windows PowerShell scripts without destroying the original cluster. The scripts allow you to dynamically change the NetNames Active Directory integration and change to and from domain joined to a workgroup and vice versa.

What is Persistent memory (PMEM)

offers a new type of memory technology that delivers a combination of capacity and persistence. Essentially, this is super-fast storage on a USB flash drive. This is deploys by using Storage Spaces Direct.

System Insights

a feature of Windows Server that provides machine learning and predictive analytics to analyze data on your servers.

What are parts of a failover cluster

Nodes, A network, a resource, Cluster Storage, Exchange Server, clients, a service or application, and a witness

In a failover cluster each node has

Full connectivity and communication with each other

Aware when other nodes leave and join the node

connects to a network through which client computers can access the clusters

Connects through a a shared bus or iSCSI connection to shared storage

Is aware of the services or applications that run locally and the resources that run on all other cluster nodes

Most clustered applications and their resources are assigned to how many nodes at once

1

Cluster storage usually refers to

logical device such as drives or logical unit numbers (LUNs) that all cluster nodes attach to through a shared bus.

A quorum is the

majority of voting nodes in an active cluster membership plus a witness vote.

Each node votes to what

to determine whether the cluster continues to run

What happens if an even number of nodes exist in a cluster

A witness votes

The cluster stops providing failover protection if

more that half of the nodes fail or if a problem occurs with communication between the cluster nodes

Without a quorum mechanism each set of nodes can

continue to operate as a failover cluster, this results in a partition within the cluster

A quorum prevents two nodes from operating

concurrently operating as a failover cluster resource

Phases to achieve quorum

After a node starts running, it determines whether other cluster members exist with which it can communicate. This process might simultaneously occur on multiple nodes. After establishing communication with other members, the members compare their membership views of the cluster until they agree on one view, based on timestamps and other information. The nodes determine if this collection of members has a quorum or enough members to create enough votes such that a split scenario can't exist. A split scenario means that another set of nodes in this cluster run on a part of the network that's inaccessible to these nodes.

Types of quorum node configurations

Node majority. Each available and communicating node can vote. The cluster functions only with a majority of the votes. This model is preferred when a cluster consists of an odd number of server nodes; no witness is necessary to maintain or achieve a quorum.

Node and disk majority. Each node plus a designated disk in the cluster storage (the witness disk) can vote when they're available and in communication. The cluster functions only with a vote majority. This model is based on an even number of server nodes being able to communicate with one another in the cluster and with the witness disk.

Node and file share majority. Each node plus a designated file share that an administrator created, the file share witness, can vote when they're available and in communication. The cluster functions only with a vote majority. This model is based on an even number of server nodes being able to communicate with one another in the cluster and with the file share witness.

No majority: In a disk-only scenario, the cluster has a quorum if one node is available and in communication with a specific disk in the cluster storage. Only the nodes that are also in communication with that disk can join the cluster.

The dynamic quorum mode does what

dynamically adjusts the quorum votes based on the number of server that are online

What is a dynamic witness

a vote that only exist if an even number on nodes exist

Types of witnesses

Witness disk - A witness disk is still the primary witness in most scenarios, especially for local cluster scenarios. In this configuration, all the nodes have access to a shared disk. One of the greatest benefits of this configuration is that the cluster stores a copy of the cluster database on the witness disk.

File share witness - A file share witness is ideal when shared storage isn't available or when the cluster spans geographical locations. This option doesn't store a copy of the cluster database.

Azure Cloud Witness - The Azure Cloud Witness is the ideal option when running internet-connected stretch clusters. This removes the need to set up a file share witness at a third datacenter location or a virtual machine in the cloud. Instead, this option is built into a failover cluster. This doesn't store a copy of the cluster database.

you can create a file share witness that utilizes a

local user account

What is cluster Performance History

A new feature in Server 2019 that gives Storage Space Direct Administrators easy access to historical compute, memory, network, and storage measurements across an organization’s servers, drives, colime, VMs, and other resources

The performance history is automatically collected and stored for up to

a year

How can you see the performance history

by using Windows Admin Center or by using the Get-ClusterPerf cmdlet

Windows Server System Insights(System Insights) is a feature that

uses Machine learning(ML) and predictive analytics (AI) tool that analyzes data to give insight into the functioning of the server deployments.

By default, System Insights offers.

CPU capacity forecasting

total storage consumption forecasting

networking capacity forecasting

volume consumption forecasting

What is the cmdlet to see System Insights

Get-InsightsCapability

Persistent Memory(PMEM) is

super fast storage on a memory sticky

Cluster set offer

cloud scale out technology, they also dramatically increase cluster node count in a single Software Defined Data Center (SDDC). This is a loosely coupled grouping of multiple failover clusters enabling VM fluidity across member clusters within a cluster set and a unified storage namespace across the set. Cluster sets preserve existing failover Cluster management experiences on member clusters and offer key use cases around forecasting lifecycle management

Windows Server 2019 now allows what of cluster nodes

in place upgrades

Microsoft Distributed Coordinator (MSDTC) now works with

Cluster Shared Volumes(CSVs)

What are some new HCI improvements in Windows Server 2019

ability to lose one node plus another drive

Windows Server 2019 clusters no longer need

NTLM authentication

In Windows Server 2019 you can now move clusters across

Domains without destroying the original cluster

In Windows Server 2019 Storage Spaces Direct not check for what before skipping them

make progress

distributed network allows for what

The use of each VMs IP instead of a dedicated one for the whole cluster

Failover Clustering is best suited for

stateful applicattions restricted to a single set of data

What are some factors to use when planning node capacity in a failover cluster

Distribute highly available applications from a failed node to all the nodes in the cluster

Ensure each node has enough capacity to handle the highly available service or application

Use similar capacity for all the nodes in the cluster

Use standby servers to simplify capacity planning

Nodes should use what type of hardware

similar

What are the infrastructure requirements for failover clustering

DNS

A domain role

An account for administering the cluster