Nutanix Cluster Initialization
I was given a Nutanix cluster with 2 blocks (8 nodes) to learn about the product, features and interoperability. During the installation and configuration process I recorded a video to demonstrate how simple and easy Nutanix really is.
In the first video of this series I demonstrated how I quick switched Nutanix Hyper-V to vSphere. This second video demonstrates how to create a brand new cluster and how Nutanix uses IPv6 Neighbor Discovery Protocol (Bonjour) to identify nodes and blocks available to initialize a cluster. Yet in this video you will have a glimpse of the beautiful Nutanix UI, called PRISM.
In my next video I will demonstrate how to configure the Nutanix cluster and create the first storage container. Keep tuned!
Watch in 720P (HD)
This article was first published by Andre Leibovici (@andreleibovici) at myvirtualcloud.net.
6 comments
Josh Sinclair
How did you get the bonjour services to actually work in Windows? What version of windows are you using / IE version? The bonjour setup has never worked for me unless I use a Mac. It is a little frustrating because my preferred method of deployment would be to just have blocks shipped out to the site and then drop a VM in the VLAN to run the cluter_init.
Andre Leibovici
Hi Josh, you need to install Bonjour Print Services for Windows.
You may download it from http://support.apple.com/kb/dl999
Josh Sinclair
I have the Bonjour services installed, they have just never worked in our environment. We're using windows 8.1 / IE 11 and I haven't had time to set up a lab and try the different versions of IE, which is why I was curious what version you are using in that demo.
Andre Leibovici
Josh, I have never tried with Windows 8.x and I was using Windows 7 for the video demo I recorded. If Bonjour is supported by Apple with Windows 8.x it should work. Maybe also try using Chrome browser.
comotoza
Am I right thay Nutanix should write Random on SSD to OpLog, Sequential to HDD. Why in none guide i've seen that i should change I/O Order?
default settings:
ncli> tier get-default-io-priority-order
Default Random IO Priority Order (High to Low):
SSD-PCIe, SSD-SATA, DAS-SATA
Default Sequential IO Priority Order (High to Low):
SSD-PCIe, SSD-SATA, DAS-SATA
ncli> ctr list
ID : XXX
Name : XXX
Storage Pool ID : XXX
Max Capacity : 11.35 TB (12,482,650,258,632 bytes)
Reserved Capacity : 0 bytes
Replication Factor : 2
NFS Subnet Whitelist :
VStore Name(s) : DemoContainer
Random I/O Pri Order : SSD-PCIe,SSD-SATA,DAS-SATA
Sequential I/O Pri Order : SSD-PCIe,SSD-SATA,DAS-SATA
Oplog Configured : true
Oplog Highly Available : true
Advanced Oplog Parameters : RF=2, Stripes=1, Synch-Write=false
Compression : on
Compression Delay : 0 mins
Fingerprint On Write : on
Andre Leibovici
comotoza,
Yes, you are correct. The OpLog is stored on the SSD tier on the CVM to provide extremely fast write I/O performance, especially for random I/O workloads. For sequential workloads the OpLog is bypassed and the writes go directly to the extent store (HDD).
Comments are preserved from the original site and are closed.