Installation & Configuration

I am new and watching all videos. I try to set up nutanix like proxmox with:3 x Dell R710 Hosts (nutanix installed on usb internal drive)1 x Dell R720 Freenas 20TB storage connected with direct Fiber With proxmox fiber channel setup was a nightmare, so can I use Fiber Channel here with Nutanix? My goal is if one of a host dies I have other VMs be ready | transferred into another host so like in VMWARE (pricy license). Let me know if I can achieve same with Nutanix and direct me towards hardware (Fiber Channel or 10GB ISCSI) whatever is easier and works well.

Hello,I am new to ovs and AHV configurations in general and have a networking question. I have been reading through the documentation on networking and still am not clear on a question or two.Situation: I am mapping out steps to create a DMZ zone for my server environment. My firewall will be configured with a DMZ port and vlan tagged, public IP will NAT to server in DMZ zone, and ACLs will control what traffic can get threw. This will be routed through vlan tagged ports via lay 3 switch to my Nutanix cluster. The server mentioned above will be on my Nutanix cluster and will need to communicate with with 1-4 other Nutanix VMs in the same DMZ/vlan.Questions: If I create a "network" in Prism with vlan tagging and apply it to all  DMZ servers in question, will this keep my DMZ traffic separated from all trusted traffic?If not, do I need to configure a new bridge to keep this network traffic separate from my other networks? My concern is because all traffic is being routed to the same trun

Foundation Central Foundation Central Overview Limitations Foundation Central OverviewFoundation Central can create clusters from factory-imaged nodes and reimage existing nodes that are already registered with Foundation Central, remotely from Prism Central. This provides benefit such as creating cluster on remote sites such as ROBO, without arranging for a deployment personnel visit.LimitationsFoundation Central has the following limitations: Foundation Central supports creation of one node cluster without imaging on factory shipped nodes with AOS and any hypervisor installed. Foundation Central supports creation of one node cluster with imaging only on nodes that are shipped out of factory with DiscoveryOS installed.Note: If you want to create a one or two-node cluster, ensure that the nodes you select support the creation of a one or two-node cluster. Imaging logs of the remote node are not available on Prism Central. These logs are only accessible within the nodes. It does

Hi Allhttps://portal.nutanix.com/page/documents/details?targetId=AHV-Admin-Guide-v5_19:ahv-cluster-nw-vs-about-ahv-c.htmlFollow the link,after the AHV was upgraded to 5.19 or lateer,  the br0 of OVS was replaced with vs0,If br1 or br2 can coexist or need to be manually migraded,does anyone have some experience .Thank you

I have been trying to chase down and eliminate this error for a while now. The cvm leader only wants to use itself as the time source - when I run this command  (allssh ntpq -pn)  it shows      remote           refid      st t when poll reach   delay   offset  jitter==============================================================================x.x.x.x  y.y.y.y    2 u  603 1024  377    0.337  55090.5 419.722x.x.x.x  184.105.182.16   3 u  200 1024    4    0.358  75064.8   0.000*127.127.1.0     .LOCL.          10 l  587 1024  377    0.000    0.000   0.000 (The asterisk indicates that it is using itself as a time source, is that correct?) I can run ntpdate successfully to the configured ntp servers and the cvm can connect to them.  (I can even run ntpdate to an external time server that is not configured for that matter)   How do I get the cvm leader to use the configured ntp servers as the source (and not use itself)?  

Hi,I followed this guide:Installing Nutanix Community Edition (CE) on vSphere 7 - Derek Seaman's IT BlogTo test a install of AHV/Prism CE and the AHV host installs fine but the CVM can not get to the outside world or ping ip’s.I can ssh to the CVM from the host on the correct IP and do the one_node_cluster create command but it just fails with `Failed to reach a node where Genesis is up` if I do `genesis start` it gets past that but just hangs after it starts up some services during the create.I tried not checking the one_node_cluster option in the install and doing it manually afterwards, but I get the same results.Happy to provide any info. The gateway etc look fine in the `/etc/sysconfig/network-scripts/ifcfg-eth0` file

We are running an image from HPE Aruba for a Mobility Master that crashes daily. The error is an IRQ interrupt:User: [14730.377720] irq 27: nobody cared (try booting with the "irqpoll" option) [14730.378230] handlers: [14730.378461] [<ffffffff813be5d0>] vring_interrupt [14730.378717] Disbling IRQ #27I’ve been directed by the community that there is a KVM fix for this by changing the machine type to Q35. I can see in the acli documentation there are settings for machine type but I can’t seem to find a list of acceptable settings that I could use there. I was hoping to perhaps create a VM through the acli with this already set or perhaps update a VM (is this bad?)Perhaps it would look something like this:acli vm.update machine_type="pc-q35"

I’m getting the following error on a 3 node clean install and I’m able to ping all 3 IPMI’s.Please advise. 

Hi Team, I will be carrying out upgrades tonight at 8 pm GMT and want to plan before hand: can I please get urgent assistance: I understand that I can jump from AOS 5.15.6 LTS to 6.0.1.1 directly but before doing this I want to confirm below: The order I can see is to upgrade Foundation 1st: we currently have foundation-4.6.2-6e9b7fc8 version can we directly jump to 5.0.4 version of foundation? or we have to install from older base versions one by one to reach 5.0.4? NCC 2nd: we currently have 4.1.0 can we jump to 4.2.0.2 or we have to install 4.2.0.1 first?

Hi everybody,In this moment, I have a 1-node cluster AHV, but in a few months this cluster will be expand could be with 1 or 2 nodes more.May I expand to 2-nodes cluster? If it is possible Is necessary a witness VM?If I have the opportunity, I think buy other 2 nodes, I think that I will not have issue expand from 1-node to 3-nodes cluster. Are I right?What do your recommend in both scenarios?

This seems like the place to post this, so please tell me if I need to move it.I am trying to setup a lab with a Nutanix G3 3060 cluster running 5.15. Using AHV, which I don’t have a lot of experience with. I think I am having a networking problem. For lab I’m only using 1GB connections.On each link I have a native or untagged VLAN which the CVM’s use. I also have several other Tagged or Trunked VLANs that I want to use for other purposes in the lab.The CMV’s are fine and the cluster comes up no problem. It has some complaints like NTP and the like, but it is otherwise good.Under networking I added the other VLANs and their VLAN IDs.I then created an OPNSense VM and gave it 2 NICs. One on the management VLAN. The same one the CVM’s use that is native. Then a second NIC with one of the tagged VLANs .I have verified that both NICs are set to connected. I even tried adding a 3rd. The management NIC (untagged) comes up and I can access the VM, but the other one won’t. OPNSense is reporting

I am trying to install ubuntu server 20.04 LTS but everytime I reach the point where to set the profile (name, username and password etc...), I get an error “Sorry, there was a problem completing the installation”…I have downloaded the ISO file from Ubuntu, and my VM settings are:CPU = 1Core or. CPU = 2Memory = 4Cd-rom = ide.0Disk = scsi.0 and 100 GiB Please advice how I can fix this, and install my Ubuntu server 20.04 LTS

HiI would like to know which is the minimun VM witness that I need for an specific scenario with several sites and 2-node cluster on each site. At the following article I see this:You can register multiple (different) Metro cluster pairs to a single Witness VM. One Witness VM can support up to 100 instances, that is the combined total number of protection domains in the registered Metro cluster pairs and the number of registered two-node clusters. (A single Witness VM can support both metro availability and two-node clusters.)https://portal.nutanix.com/page/documents/details?targetId=Prism-Element-Data-Protection-Guide-v5_18:sto-metro-availability-witness-c.html However at the following post I see this: You can register multiple (different) Metro cluster pairs to a single Witness VM. One Witness VM can support up to 50 Witnessed Metro protection domains distributed among its registered Metro cluster pairs.https://next.nutanix.com/installation-configuration-23/metro-availability-witness

I have deployed a three node Nutanix cluster (version 5.18) in VMware ESXi. I am trying to deploy Prism Central on the AOS but while uploading the json and tar file of the prism central to the Nutanix cluster, I am getting the error ‘unable to upload file’ at around 6% and the upload fails.Kindly help me to sort out this issue.

I am attempting to automate the data-at-rest encryption setup on our clusters and I am struggling with the certificate upload api.  Does anyone have examples of using this api or could help me figure out how to make it work?  Prism V2 APIPost:/data_at_rest_encryption_certificates/svm_certificates/multi_file {  "certificate_map": {},  "key_management_server_name": "string"} I am not sure what to enter for the certificate_map.  Is the the contents of the certs, a file, or something else?   

Hello everybody, I searched everychere but I didn’t find what I wanted. I will explain you my problemI need to update the disk on a VM but I have about 100 VM and I don’t want to update each one by one with GUI on the web interface of the Prism Central interfaceI want to know what is the best way to do it. I think that is possible with acli but I don’t find command to do it or with API but with it I don’t know if i’s possible but I guess that it isThank to all for helping me

MetisMetis ToolMetis allows you to qualify and integrate your hardware with Nutanix software.Nutanix delivers Metis as an .ova or QCOW2 file depending on the hypervisor. Deploy it on a separate utility node or on a laptop using VirtualBox.Note: If you host Metis on a laptop, make sure that the laptop stays running and connected throughout the entire process.After using Metis, submit the results to the Nutanix Elevate portal. Nutanix uses the results to assess whether the individual hardware components or the entire platform configurations meet the standards to support Nutanix software.Infrastructure Requirements Foundation VM X-Ray VM DHCP server Elevate PortalDetails of the Nutanix Elevate portal.Nutanix maintains a web portal called the Elevate portal as a central location for FleX information, which includes Metis.After you run tests on your component or platform using Metis, you can upload the results to the Elevate portal.When you create a qualification ticket in the Elevate p

OverviewFleX Driver Development KitDriver development overview.About this taskNutanix provides the FleX Driver Development Kit (DDK) to help developers build, inject, qualify, and distribute new drivers.Use the DDK to build a Linux kernel with one or more loadable kernel modules. The DDK builds a Linux kernel tar file that you can use as an input for Metis.In general, the driver development procedure is as follows:Procedure Prepare the compile environment. Prepare the source code to compile. The source can be an updated kernel source tree or one or more out-of-tree kernel module. Compile the kernel and modules to create the kernel tar file with the driver. Use the new kernel tar file as an input for Metis. Use Metis to qualify the new kernel and driver. Submit the driver source code (source rpm), qualification results, the procedure for building the driver, and release notes though your qualification ticket on the Elevate portal. FleX DDK RequirementsDriver development requir

Hello!I tried to create a user with the backup role and get an error “Error saving user roles. The specified role name ROLE_BACKUP_ADMIN is invalid.”New installed cluster: 3nodes Xeon/128GB/2x1TB SSD/240GB SSDHelp me, pls! 

I have an end-of-life device and the support portal is out of service. How can I access the files I need to download?

Can anyone tell me about the nutanix license If pro license expired. Will there be assistance when opening the ticket?And the license status is pro or starter?Thanks

The instructions for shutting down a cluster are to use “sudo shutdown -P now” on the CVM’s in the cluster to shut each one down, but the command only works on the first CVM.   The command errors out the rest of the CVMs because they can’t reach the first CVM that was shut down.   Am I missing something?

we had 4 finding for pci dss compliance on AOS node :1. 57608 - SMB Signing not requiredSigning is not required on the remote SMB server.5.3 (CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N)tcp port 4452. 84502 - HSTS Missing From HTTPS ServerThe remote web server is not enforcing HSTS.6.5 (CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N)tcp port 59893. 51192 - SSL Certificate Cannot Be TrustedThe SSL certificate for this service cannot be trusted.6.5 (CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N)tcp port 59894. 57582 - SSL Self-Signed CertificateThe SSL certificate chain for this service ends in an unrecognized self-signed certificate.6.4 (CVSS2#AV:N/AC:L/Au:N/C:P/I:P/A:N)tcp port 5989may i know how to solve it ?

Hi Everyone, My Nutanix cluster had 4 node running AHV version Nutanix 20170830.337 and AOS version 5.10.9.1 LTS.Suddenly, one of the node had CVM unreachable issue. When I’m running cluster status command got message 1 CVM down.I’m try to run nodetool -h 0 ring got something like this. When I’m issue cmd ncli host list. I’m notice the problem node got this message “Node marked to be added back to metadata store”1 of my colleague try to repair host boot device but no luck. What should we do now?  

I need to upgrade an dark site cluster from AOS 5.1.5 to 5.20.1.1.   It looks like I need to upgrade from 5.1.5 to 5.5.9.5 to 5.10.11.1 to 5.15.6 to 5.20.1.1.   Will I need to upgrade the AHV matched to each version after I upgrade the AOS?