I’ve always been interested in investigating VMware Site Recovery Manager but never had the time nor the kit to work through it in my lab. That all changed recently whith 2 developments:
- I acquired a secondhand server with dual 8-core processors, 48GB of RAM and some decent 10K SAS drives.
- VMware released SRM 8.2 which is distributed as a pre-prepared appliance, making deployment incredibly easy.
So let’s see how I did it. Read on…
Note: this is part 1 of a 10 part series. Links to the rest of the series can be found at the bottom of the page.
The first thing to say is that my lab deployment isn’t meant to represent best practice for a production environment but simply shows you how to get SRM up and running. As with all technologies, plenty of reading of best practices documentation is a must.
The first step is to show you how I intend to build out my virtual infrastructure. The physical server is running an evaluation copy of vSphere ESXi 6.7. This forms the basis of my virtual lab. I have 4 virtual switches, one with an uplink connecting everything to the internet via my home router as shown in the schematic below. The virtual switches are connected to each other using a router based on the VyOS virtual router version 1.1.8. You can download an OVA version of this here and there are installation instructions here. You’ll need to edit the properties of the deployed VM before you boot it up with the correct number of NICs and connections to the virtual switches.
(Schematic created using https://www.draw.io/)
The commands I’ve used to configure my VyOS router are as follows:
#set interfaces ethernet eth0 address 192.168.1.250/24
#set interfaces ethernet eth1 address 10.0.0.254/24
#set interfaces ethernet eth2 address 10.0.1.254/24
#set interfaces ethernet eth3 address 10.0.2.254/24
#set nat source rule 10 outbound-interface eth0
#set nat source rule 10 translation address masquerade
#set system gateway-address 192.168.1.254
#set system name-server 8.8.8.8
I also like to configure DHCP relay on my VyOS router. My dhcp server is attached to the 10.0.0.0/24 network and I’d like to use it to distribute IP addresses to the 10.0.1.0/24 and 10.0.2.0/24 networks. This is a simple process of entering the following commands:
#set service dhcp-relay interface eth1
#set service dhcp-relay interface eth2
#set service dhcp-relay interface eth3
#set service dhcp-relay server 10.0.0.7 (this is the local DNS server)
#set service dhcp-relay relay-options relay-agents-packets discard
Enter the following command to save the setup:
#commit
#save
I’ve prebuilt a Windows Server Domain Controller with DNS and DHCP roles installed. This will be my main management station for deploying all the components in the solution. You’ll notice that I have configured this system with 2 NICs and connected it to both the 192.168.1.0 and 10.0.0.0 networks. The reason I’ve done this is so I can use it as a DHCP server on the 10.0.0.0 network (but not for this particular lab) without interfering with the DHCP offered by my home router. It has a default gateway of 192.168.1.254 i.e. the address of my home router. In order for it to find the systems attached to other networks I’ve created static routes as follows:
c:\>route -p ADD 10.0.1.0 MASK 255.255.255.0 10.0.0.254
c:\>route -p ADD 10.0.2.0 MASK 255.255.255.0 10.0.0.254
My preferred browser is the current release of Chrome. I’ll mainly be using the new HTML5 vSphere client but for some things I still needed to use the older Flash-based client so I’ve also installed the latest version of Flash Player.
I’ve created DNS entries for all the systems I’ll be deploying as follows:
| System | Function | IP Address |
| dc-01 | Domain Controller, DNS, DHCP, NTP, management | 10.0.0.7 |
| vRouter | Vyos router | 10.0.0.254 |
| vcsa-prod-01 | vCenter Server, Production | 10.0.1.1/24 |
| vrep-prod-01 | vSphere Replication Appliance, Production | 10.0.1.2/24 |
| srm-prod-01 | SRM Appliance, Production | 10.0.1.3/24 |
| vfas-prod-01 | NetApp Storage Appliance, Production | 10.0.1.4/24 |
| vsc-prod-01 | NetApp Virtual Storage Console, Production | 10.0.1.5.24 |
| esx-prod-01 | Nested ESXi Host, Production | 10.0.1.11/24 |
| vcsa-dr-01 | vCenter Server, DR | 10.0.2.1/24 |
| vrep-dr-01 | vSphere Replication Appliance, DR | 10.0.2.2/24 |
| srm-dr-01 | SRM Appliance, DR | 10.0.2.3/24 |
| vfas-dr-01 | NetApp Storage Appliance, DR | 10.0.2.4/24 |
| vsc-dr-01 | NetApp Virtual Storage Console, DR | 10.0.2.5.24 |
| esx-dr-01 | Nested ESXi Host, DR | 10.0.2.11/24 |
Don’t worry about the NetApp stuff for now – that’s for some future testing I’ll be doing.
I’ll be using the following versions of the software, all of which will be deployed in evaluation mode so no need to obtain licenses:
vCenter Server 6.7
vSphere ESXi 6.7
vSphere Replication 8.1
Site Recovery Manager 8.2
All of these can be downloaded from the vmware.com website.
So that’s my lab environment. Yours maybe different. The limiting factor is memory – I ended up using about 40GB of memory – but if you have SSDs, overcommiting becomes less problematic as you will be swapping out to flash. You may also consider spreading the lab across 2 physical servers. Whichever path you choose, hopefully you’ll find the following series of posts helpful.
Part 1 – Overview
Part 2 – Nested ESXi Hosts
Part 3 – Deploy vCenter Servers
Part 4 – Deploy vSphere Replication
Part 5 – Configure vSphere Replication
Part 6 – First Replication Job
Part 7 – Deploy SRM Appliance
Part 8 – Configure SRM – Site Pairing
Part 9 – Configure SRM – Mappings
Part 10 – First Recovery Plan
Tidak ada komentar:
Posting Komentar