Veritas InfoScale™ 7.3.1 Virtualization Guide - Linux on ESXi

Last Published:
Product(s): InfoScale & Storage Foundation (7.3.1)
  1. Section I. Overview
    1. Overview of Veritas InfoScale solutions in a VMware environment
      1.  
        Overview of the Veritas InfoScale Products Virtualization Guide
      2. Introduction to using Veritas InfoScale solutions in the VMware virtualization environment
        1. How Veritas InfoScale solutions work in a VMware environment
          1.  
            How Veritas InfoScale product components enhance VMware capabilities
          2.  
            When to use Raw Device Mapping and Storage Foundation
          3.  
            Array migration
          4.  
            Veritas InfoScale component limitations in an ESXi environment
          5.  
            I/O fencing considerations in an ESXi environment
      3. Introduction to using Dynamic Multi-Pathing for VMware
        1.  
          About the SmartPool feature
      4.  
        About the Veritas InfoScale components
      5. About Veritas InfoScale solutions support for the VMware ESXi environment
        1.  
          Veritas InfoScale products support for VMware functionality
      6.  
        Virtualization use cases addressed by Veritas InfoScale products
  2. Section II. Deploying Veritas InfoScale products in a VMware environment
    1. Getting started
      1.  
        Veritas InfoScale products supported configurations in an VMware ESXi environment
      2.  
        Storage configurations and feature compatibility
      3.  
        About setting up VMware with Veritas InfoScale products
      4.  
        Veritas InfoScale products support for VMware environments
      5.  
        Installing and configuring storage solutions in the VMware virtual environment
  3. Section III. Use cases for Veritas InfoScale product components in a VMware environment
    1. Storage to application visibility using Veritas InfoScale Operations Manager
      1. About storage to application visibility using Veritas InfoScale Operations Manager
        1.  
          About Control Hosts in Veritas InfoScale Operations Manager
      2. About discovering the VMware Infrastructure using Veritas InfoScale Operations Manager
        1.  
          Requirements for discovering vCenter and ESX servers using Veritas InfoScale Operations Manager
        2.  
          How Veritas InfoScale Operations Manager discovers vCenter and ESX servers
        3.  
          Information that Veritas InfoScale Operations Manager discovers on the VMware Infrastructure components
        4.  
          About the datastores in Veritas InfoScale Operations Manager
        5. About the multi-pathing discovery in the VMware environment
          1.  
            About the user privileges for multi-pathing discovery in the VMware environment
        6. About near real-time (NRT) update of virtual machine states
          1.  
            Setting-up near real-time (NRT) update of virtual machine states
          2.  
            Configuring the VMware vCenter Server to generate SNMP traps
      3.  
        About discovering LPAR and VIO in Veritas InfoScale Operations Manager
      4.  
        About LPAR storage correlation supported in Veritas InfoScale Operations Manager
    2. Application availability using Cluster Server
      1.  
        About application availability with Cluster Server (VCS) in the guest
      2.  
        About VCS support for Live Migration
      3.  
        About the VCS for vSphere setup
      4.  
        Implementing application availability
      5.  
        Assessing availability levels for Cluster Server in the VMware guest
    3. Multi-tier business service support
      1.  
        About Virtual Business Services
      2.  
        Sample virtual business service configuration
    4. Improving storage visibility, availability, and I/O performance using Dynamic Multi-Pathing
      1.  
        Use cases for Dynamic Multi-Pathing (DMP) in the VMware environment
      2.  
        About Dynamic Multi-Pathing for VMware
      3. How DMP works
        1. How DMP monitors I/O on paths
          1.  
            Path failover mechanism
          2.  
            I/O throttling
          3.  
            Subpaths Failover Group (SFG)
          4.  
            Low Impact Path Probing (LIPP)
        2.  
          Load balancing
        3.  
          About DMP I/O policies
      4.  
        About storage visibility using Dynamic Multi-Pathing (DMP) in the hypervisor
      5.  
        Example: achieving storage visibility using Dynamic Multi-Pathing in the hypervisor
      6.  
        About storage availability using Dynamic Multi-Pathing in the hypervisor
      7.  
        Example: achieving storage availability using Dynamic Multi-Pathing in the hypervisor
      8.  
        About I/O performance with Dynamic Multi-Pathing in the hypervisor
      9.  
        Example: improving I/O performance with Dynamic Multi-Pathing in the hypervisor
      10.  
        About simplified management using Dynamic Multi-Pathing in the hypervisor and guest
      11.  
        Example: achieving simplified management using Dynamic Multi-Pathing in the hypervisor and guest
    5. Improving I/O performance using SmartPool
      1.  
        Improving I/O performance with Veritas InfoScale product components in the VMware guest and DMP for VMware in the ESXi host
      2.  
        Implementing the SmartIO and SmartPool solution
    6. Improving data protection, storage optimization, data migration, and database performance
      1.  
        Use cases for Veritas InfoScale product components in a VMware guest
      2. Protecting data with Veritas InfoScale product components in the VMware guest
        1.  
          About point-in-time copies
        2.  
          Point-in-time snapshots for Veritas InfoScale products in the VMware environment
      3. Optimizing storage with Veritas InfoScale product components in the VMware guest
        1.  
          About SmartTier in the VMware environment
        2.  
          About compression with Veritas InfoScale product components in the VMware guest
        3.  
          About thin reclamation with Veritas InfoScale product components in the VMware guest
        4.  
          About SmartMove with Veritas InfoScale product components in the VMware guest
        5.  
          About SmartTier for Oracle with Veritas InfoScale product components in the VMware guest
      4. Migrating data with Veritas InfoScale product components in the VMware guest
        1.  
          Types of data migration
      5. Improving database performance with Veritas InfoScale product components in the VMware guest
        1.  
          About Veritas InfoScale product components database accelerators
      6.  
        Simplified storage management with Veritas InfoScale product components in the VMware guest
    7. Setting up virtual machines for fast failover using Storage Foundation Cluster File System High Availability on VMware disks
      1.  
        About use cases for Storage Foundation Cluster File System High Availability in the VMware guest
      2.  
        Storage Foundation Cluster File System High Availability operation in VMware virtualized environments
      3.  
        Storage Foundation functionality and compatibility matrix
      4. About setting up Storage Foundation Cluster File High System High Availability on VMware ESXi
        1.  
          Planning a Storage Foundation Cluster File System High Availability (SFCFSHA) configuration
        2.  
          Enable Password-less SSH
        3.  
          Enabling TCP traffic to coordination point (CP) Server and management ports
        4. Configuring coordination point (CP) servers
          1.  
            Configuring a Coordination Point server for Storage Foundation Cluster File System High Availability (SFCFSHA)
          2.  
            Configuring a Coordination Point server service group
          3.  
            Configuring a Cluster Server (VCS) single node cluster
        5.  
          Deploying Storage Foundation Cluster File System High Availability (SFCFSHA) software
        6.  
          Configuring Storage Foundation Cluster File System High Availability (SFCFSHA)
        7.  
          Configuring non-SCSI3 fencing
      5. Configuring storage
        1.  
          Enabling disk UUID on virtual machines
        2.  
          Installing Array Support Library (ASL) for VMDK on cluster nodes
        3.  
          Excluding the boot disk from the Volume Manager configuration
        4.  
          Creating the VMDK files
        5.  
          Mapping the VMDKs to each virtual machine (VM)
        6.  
          Enabling the multi-write flag
        7.  
          Getting consistent names across nodes
        8.  
          Creating a clustered file system
  4. Section IV. Reference
    1. Appendix A. Known issues and limitations
      1.  
        Prevention of Storage vMotion
    2. Appendix B. Where to find more information
      1.  
        Veritas InfoScale documentation
      2.  
        Service and support
      3.  
        About Veritas Services and Operations Readiness Tools (SORT)

Assessing availability levels for Cluster Server in the VMware guest

In certain situations, VMware HA with VCS-in-guest would yield higher availability than a standalone VCS-in-guest.

For example:

Three ESXi hosts H1, H2, and H3 and two virtual machines (VMs) N1 and N2 are in a VCS-in-guest cluster.

N1 is running on H1 and N2 is running on H2. N1 is running the application.

If H2 fails, then VMware HA restarts N2 on H3.

Now if H1 fails, the application would fail over to N2. In absence of VMware-HA this would not be possible.

In certain situations, VMware HA with VCS-in-guest would yield lesser availability than standalone VCS-in-guest.

For example:

Two ESXi hosts H1 and H2 and two VMs N1 and N2 are in a VCS-in-guest cluster.

N1 is running on H1 and N2 is running on H2. N1 is running the app.

If H2 fails, then VMware HA restarts N2 on H1. N1 and N2 end up on the same host.

Subsequently if H2 is back again and if H1 fails, both N1 and N2 fail and from the point of view of VCS-in-guest cluster, it would be as if the entire cluster rebooted.

Note:

This would yield lesser availability than stand-alone VCS-in-guest where when H2 is back again and N2 would have started on H2, thus allowing the application to failover quickly to N2 when H1 fails.

The goal is that VMs of a VCS-in-guest cluster must distribute themselves evenly across hosts, otherwise there is a possibility of losing availability. It is difficult to enforce this through DRS because DRS triggers only periodically, which leaves a window where VMs may not be distributed evenly.