VMware vCenter SRM: Inventory Mapping

we're working our way through the deployment of SRM the next piece of the puzzle is what we call inventory mapping in inventory mappings if you read through the documentation are entirely optional this is not a required element you do not have to do inventory mappings but I hope to deploy or to explain to you in the next little while why there's something that you would really want to do as a part of your SRM deployment so inventory mappings are relatively straightforward think about your environment this way you've got a virtual machine that you'd like to protect that lives for instance in your New York data center when the New York data center has a problem and somebody decides that a failover needs to occur where is that virtual machine going to show up in the Chicago data center because that's its designated recovery site and when I say show up well what do you mean show up well don't forget a virtual machine has a home in the hosts and clusters view in other words what resource pool does it belong to also when you go to virtual machines and templates view you end up putting a virtual machine into a virtual machine folder and it's reasonable to say that you probably don't want to mix those up with the virtual machines that are running in Chicago we would like to keep those that are failing over in a unique inventory environment and what this allows us to do what inventory mappings let us do is they let us take those objects so anything that lives in a resource pool called 1 2 3 is going to failover into a resource pool called 4 5 6 or something like that I mean your naming is completely up to you the mapping is what ties the two together we do exactly the same thing with networks so it's reasonable to say that I've got a production network in the New York office but I maybe don't want my virtual machines when they failover from New York over to Chicago to actually show up on the Chicago production network we may want them on an isolated network into themselves we may need them on a completely different VLAN and the way we're going to do that is we're going to deploy what we call inventory mappings which are essentially placeholders which say when this view machine fails from the protected site to the recovery site we're going to place him on this network and that information is all stored in what we call it's a configuration file for the virtual machine when it fails over and those need to be stored in what we call a placeholder data store so as I said earlier when we introduced this module if the virtual machine has one of these elements which is not completely mapped then when you go to protect the virtual machine when you go to list that VM for protection you're going to be required to go in and solve that problem in other words I'd like to protect my database server in New York if I don't already have his network mapped then when you go to protect it it will ask you it will come up and prompt you and I don't know if you're kind of like me and I'm going to protect fifty seventy five a hundred virtual machines then I don't want to have to go through that manually I'd like to say do this and from there on in if I protect a VM that's connected to this network it will automatically be connected to that network when it fails over and the same principle applies to virtual machine folders and also resource pools okay so the demonstration that we're about to embark on is going to show you exactly this we're going to walk through configuring inventory mappings so we're going to look at the resource pools we're gonna look at virtual machine folders we're going to look at the network mappings we're also going to configure those placeholder data stores and then last but not least I'm also going to configure our VMs for protection using vSphere replication so now that we've got the entire environment deployed we're now going to say I would like to protect this virtual machine so that the replication process starts so that we can then move forward with the next step in actually deploying and configuring SRM and really now we're more in the configuring stage so inventory mapping has the subject on the table right now so if we select our site go to the vSphere replication dialog box we can see that we're finished with all of that stuff we got the green checks beside all of those following on from the last piece go back to our sites dialog box and if we look down the list of items on the getting started tab we see the ability to set up inventory mappings so click on one those links go through resource mappings for instance and you'll notice that we've now browsed into the environment where we show all of the virtual machine resource pools that are listed as a part of the honey-do o3 site so what we're going to do is we're going to configure an environment whereby we create a mapping for the protected services over to the recovered services on the opposing site so in other words when a virtual machine that lives in the application servers folder or application servers resource pool on the protected site actually fails over when it goes through the failover process it will show up on the recovery site under a folder called recovered services which means that it won't get intermixed with all of the other running virtual machines that may be in existence on our recovery site so we simply proceed to map through all of the various different resource pools in this particular instance to go across now that's one page and those are resource mappings as we show at the top tab so once we're happy with those and by the way don't forget to take some time to have a look at those to make sure that you got them in the right place did you fail them over to the recovered services area you know and in this case it looks like we've got one that's a little interesting we've got protected services failing over two protected services and not sure that's exactly what we wanted so to fix that is simply a double click and on that dialog box or configure mapping again if you choose to do so and then we can choose the appropriate folder so you know minor items are easy to fix and the good news is there we go showing you the fix so at that point the rest looks pretty good we can proceed onwards and take a look at some of the other items recovered services here are not shown as being mapped but that's no big deal once we select folder mappings we're going to do exactly the same thing as we did under resource mappings take any folder that has VMs in it that you expect to fail over and tell SRM into which folder you want those VMs to show up should a failover occur so here we're mapping protected services to recovered services in the vm's and templates of you okay and we would do exactly the same thing on site 4 notice that we just selected site four go back and do exactly the same thing in equal and opposite direction and the idea of being here that we're building out an infrastructure whereby one site is the recovery site for the opposite site and vice versa and then network mappings it's much the same idea take any of your resource arts or your pork groups and take those and map those to the appropriate pork group on the recovery site so if a VM is plugged into prod net on Honey duo 4 we want it to be plugged into prod net on honey duo 3 and it doesn't have to be pregnant it could be some completely different network that's up to you in terms of how you do the failover process you may want to have it failover onto a completely separate and isolated VLAN but you would need to have a pork group in existence to make that happen so these pork groups have to exist before you do this task and there we've shown you doing both sides of it both from site 3 and site force perspective place holder data stores are where we put the virtual machine configuration files on the recovery site in anticipation of that VM failing over and what's stored in there it's it's a mini config file for each VM and what gets put in there are these mappings right essentially which which resource pool which folder and all the rest of those settings network connections and all those that go into the back end so we're going to simply choose a datastore which exists in the recovery site where the placeholder VMs are going to be put and the as there as we prepare to fail them over from the protected site so now that we've completed the inventory mappings and we've configured some virtual machines for protection via vSphere replication we can then start to look at the next steps however the next step for you might need to consider getting some more training on SRM and the way we do that go to vmware calm slash education search for SRM 5 install configure manage and that'll be the class that will give you all of the hands-on necessary to deploy this in your own organization

Channel: VMware

Views: 24,525

Keywords: VMware, SRM, vCenter, vCenter Site Recovery Manager, Site Recovery Manager, Disaster Protection, Getting Started with SRM, Inventory Mapping

Id: NXMCRc0ly-8

Length: 8min 54sec (534 seconds)

Published: Tue Oct 02 2012