vRealize Automation 7 – Reservations
January 25, 2016vRealize Automation 7 uses the concept of reservations to grant a percentage of fabric group resources to a business group. To add a reservation go to Infrastructure –> Reservations. Click the “New” button to add a reservation and then select the type of reservation to be added. Since I’m using a vSphere Cluster, I selected Virtual –> vCenter. Depending on what kind of reservations you’ve selected, the next few screens may be different, but I’m assuming many people will use vSphere so I’ve chosen this for my example.
Enter a Name for the reservation and the tenant (which should already be selected). Next, in the dropdown select your business group that will have access to the reservation. Leave reservation policy empty for now but enter a priority. If a business group has access to more than one reservation, the priority is used to determine which to use up first. Lastly, select “Enable this reservation”. Click “Next”.
On the resources tab, select the compute resource and then we need to add some quotas. Quotas limit how large the reservation will be, so we can limit it by a number of machines, the amount of memory or how much storage is being used. Be sure to enter a memory amount and at least one datastore to be used for deploying cloud resources. Click “Next”.
On the network tab, select the networks that can be used to deploy resources and for now leave the “Network Profile” blank.The bottom section is used with NSX or vCNS but we’ll leave that for another post.
On the properties tab, you can add custom properties that will be associated with all catalog items deployed through this reservation. For now we’ll leave this empty. Click “Next”.
Lastly, the alerts page we can set the thresholds on when to alert our administrators about resource usage.
Summary
Reservations are how we limit our business groups to a certain amount of resources in our cloud. They are necessary to prevent our vSphere environment from being over provisioned with virtual machines and can empower business group managers to handle their own resources instead of the IT Administrators.
I’ve been trying this out in my lab, and for some reason I can’t see any network are storage paths. I’ve triple checked that both ESXi hosts are connect to shared storage. Is this something you have encountered?
Do you see your clusters though? Many times the discovery has issues, but if you can see clusters that shouldn’t be the problem. Be sure to run a Discovery if you make any changes to the infrastructure or there will be a lag.
Thanks, I can see clusters, but it doesn’t see any physical RAM, networks, or storage.
I’m going to check with a co-worker and I’ll post back if I find something obvious. I’ve been poking around a bit, but do you know where discovery logs are kept?
You should see some info under infrastructure –> monitoring. I’d check there first.
Thanks,
I’m still working on what is wrong, but I manually installed a new IAAS box which seemed to fix the issue. I will keep looking at this and post back if I can figure out the cause.
So far I haven’t found a settings difference in the 2 configs, but every time I try the express installer against 2 separate vSphere 6.0u2 infrastructures, it doesn’t seem to work with the symptoms described above.
Hi,
i had the same issue like described above. No memory storage etc.
Finally i had to configure MSDTC on my SQL server for remote calls. This solved my problem…buth nowhere documented.
For those that are still having issues after uninstalling and reinstalling MSDTC here’s a potential fix. It was a head scratcher for me for about a day. If your SQL server is different than your IaaS server make sure to match the security on the MSDTC service for both servers.
http://www.sovsystems.com/vcac-6-1-and-msdtc-nuances/
I just came back to post exactly what Clemens said. I noticed when I tried to do a manual install and I needed to change the DTC settings. I ran in to a FW issue as well, but after opening TCP/135 and using the dtcping utility to test it fired right up.
I don’t see anything in the reservations section using vRA 7.0.1
Is it coz my vcenter creds are bad? but that cannot be because it did see the “cluster” name and i am using administrator@vsphere.local which has admin privileges in vcenter.
No nsx at my end
Please advise
I’d need to know more about your environment, but you create a Fabric Group first, and then you can create your reservations. Be sure that you add your login to the Fabric Administrators when you setup the Fabric Group.
Does anyone know, how to move VM from one Business Group to anothe.
Hi,
I have two questions:
1. How to reserve/allocate CPU just like memory say 20 vCPU.
2. While we dont configure CPU reservation but we can define CPU based capacity alerts? where does the number comes from?
1. As of 7.4 you aren’t able to reserve or allocate CPUs in vRA. I believe VMware assumes that you’ll always be able to overallocate CPUs so it isn’t an issue. If you need to reserve CPUs, you can use a vRO workflow to reserve CPU in vCenter through a “VCENTER” limit or reservation but not in vRA.
2. vRA does track your CPUs even though you don’t reserve or allocate them. This is where the alerts come from. This is the same as memory allocations.
Hello Eric,
How to map a reservation policy to a compute resource ??
I have created reservation policy, updated the same in reservation and blue print. In this case If have multiple compute resources ( clusters in VC ) how the vm will be deployed in the respective cluster based on reservation ? where to map the reservation policy to vcenter cluster ??
You would add the reservation policy to a reservation, and a blueprint. This effectively ties them together.