In case you loose data due to any reason you have to recover it from backup. Fortunately, you have more than one way to restore data and bring service back up into Production. In this post, we will discuss how to restore VM using one of the methods available. In the previous post, I have performed…
HyperV, SCVMM, Windows servers hosting VMs – these are some of those things that strike disbelief, apprehension and that general feeling of – nah, it’s horrible, or it’s painful to maintain. Fortunately, it’s not all so bad. Having worked on one of the largest deployments of HyperV and SCVMM in the APAC region I think…
I believe the best method to learn any technology is to teach someone or implement it. I prefer to deploy and try out for myself. Having said that I’m open to teaching if required. I have recently started exploring the feasibility of using Windows Server 2016 Nano for the Production environment. There are at least…
Nutanix prism allows you to expand cluster in 1-click fashion which will allow to discover, image (foundation), configure network and expand cluster (expand storage pool) in two page wizard, but some times due to unknown reason or restrictions you cannot use prism wizard..
- Use field guide to Image the node with your choice of hypervisor.
- Use phoenix to get CVM Configured on the node..
but recently you don’t get to download phoenix iso from support but instead generate from any existing CVM as below;
allssh ls ~/foundation/nos/
to find where you have AOS tar uploaded already, then ssh to that CVM..and run
it might possibly exists.. change to foundation binary directory
run below command to create phoenix iso..
./foundation --generate_phoenix --nos_package=/home/nutanix/foundation/nos/ nutanix_installer_package-release-euphrates-x.x.x-stable.tar --temp_dir=/home/nutanix/foundation/tmp
replace the currect aos image tar file and version..
WinSCP to CVM and download the Phoenix ISO file to be used for New node configuration.
Once node is ready, use Prism Cluster Expand to configure. if you cannot you can always manually set IP address to hypervisor and CVM using IP v6 or through hypervisor internal network. Once IP addresses are set and connectivity is verified run below command on your Nutanix cluster to discover the new node by your Nutanix cluster.
ncli cluster discover-nodes
Which should show output as below
Cluster Id : Controller Vm Address : 192.168.29.234 Current Network Interface : eth0 Foundation Version : foundation-3.5-7ca0fb1f Hypervisor : hyperv Hypervisor Address : 192.168.29.238 Hypervisor Version : Microsoft_Windows_Server_2012_R2_Datacenter Ip : fe80::215:5dff:fe09:8c00%eth0 Ipmi Address : 192.168.29.242 Node Position : D Node Uuid : 50d9b2f9-d099-4a7f-bc4b-4417d4a4413e Nos Version : 5.0.1 Rackable Unit Model : NX-1065-G4 Rackable Unit Serial : 16SM6501XXXX Sed Node : false Svm Ip : 192.168.29.234
Note the node Uuid and run below command to add the node to cluster.
ncli cluster add-node node-uuid="50d9b2f9-d099-4a7f-bc4b-4417d4a4413e"
Go have coffee and when you are back login to Prism to see the newly added node or new storage size..
One of the life saving feature of Nutanix platform is native data protection, now every other vendor has it so whats so great about it… I do remember at one of our customer had failed upgrade of application in a critical virtual machine and to get the virtual machine out and running from storage snapshot how much it takes..
The best part of Nutanix data protetion is virtual machine granularity, it does not only give per virtual machine protection, it also enables to spin-up selective virtual machine as test/dev isolated environment in minutes.
Addition to that Nutanix have hidden feature of data protection to keep few expired snapshots in case you does not noticed you have curruption days over your retention. By default the value is keep 5 snapshots (kb1531) which can be seen from ncli as below;
ncli pd ls-sched name="pd-name"
and can be updated as below if required.
ncli pd set-schedule min-snap-retention-count=3 name="pd-name"
So set it with cuation if you have less then 5 snapshot retention and running low on disk space.
But what if the snapshot has been expired and believe me it has saved me coulpe of time when someone from top c level mess up with data and after strugling for few days to realize your snapshot retention has passed.
so what you do .. connect to Nutanix cluster by ssh..
ncli pd ls
to list all your protection domains configured..
ncli pd ls-snaps name="pd-name"
to list all the snapshots.. which show output similar as below
ID : 96731 Protection Domain : CRM-PD Create Time : 02/11/2017 23:30:00 UTC Expiry Time : 02/12/2017 23:30:00 UTC Virtual Machine(s) : 1 VM Name : CRM-VM01 Consistency Group : CRM-VM01 Application consistent... : false Power state on recovery : Powered On NFS Files : 0 Located at Remote Site : Local Snapshot Size : 15.54 GiB (16,689,091,848 bytes) ID : 96530 Protection Domain : CRM-PD Create Time : 02/10/2017 23:30:00 UTC Expiry Time : Expired Virtual Machine(s) : 1 VM Name : CRM-VM01 Consistency Group : CRM-VM01 Application consistent... : false Power state on recovery : Powered On NFS Files : 0 Located at Remote Site : Local Snapshot Size : 8.95 GiB (9,609,154,312 bytes)
there you see you have a snspshot you need on the date but Expiry Time is Expired, worry not, my retention was just 1 day an minimium retention was set 2 so Acropolis is going to retain minimium 2 snapshots.. now prism also shows expired snapshots and allows operations on it.. but i still run old NOS..for some unknown reason 😦
ncli pd restore-snapshot name="CRM-PD" snap-id="96530" vm-names="CRM-VM01" ' replace="false" path-prefix="/Restored-CRM-VM01" vm-name-prefix="Restored-"
and voila..! vm restored and resgitered in no time.. just power on and have a coffee..
If you haven’t read my previous post about the Vembu’s big free forever announcement, make sure you check it out! For small businesses or lab environments, having an option for forever free backup software is really cool. For medium and large enterprises, I’m sure that, if they try Vembu (like I did below), they will…
HyperConvergence is a hot market and more technology vendors are trying to get their share in this fastest-growing segment, either with homegrown solutions or technology acquisitions. According to Gartner the overall market for integrated systems will reach $5 billion by 2019. It’s not surprising that with these many entrants in this game it is not…