Installation & Configuration

I’m having trouble with network connectivity on a new installation of CE. The FAQ on this site says to login as root and run a command to repair the installation but I can’t find a listing for the root password.

I have 4 node, what is the best practice to connect node with production switch connect the 2 ports 10g for each node to switch and keep 1g without any connect? 

I am attempting to Install CE on a HPE Proliant DL380Gen10 server with 128GB RAM, two Xeon 4116 cpus, one 480 SSD, and four 600Gb HDD’s I created a USB install drive using Rufus and the newest version of CE. I get all of the way through setting up the drives and networking and the process appears to be going well until the part way through the ‘Hypervisor installation in progress’, then fails with Mount /dev/None does not exist message. 

Hello, I’m trying to install CE on HPE Bladesystem C7000 and installtion getting terminated  quickly. The blade had 2 disk of 150Gb in RAID0  I would like to know i’m missing anything here.  Regards,Harshal

OpenSSH_7.4p1, OpenSSL 1.0.2k-fips  26 Jan 2017 version is using in the latest LTS AOS and AHV (AOS 6.5.6.6 , AHV el7.nutanix.20220304.511 )Please help us to resolve this issue.It is recommended to use the newer version of the Paramiko Python sshd 2.4.2 module. It is recommended to disable port 2222 used by Paramiko from within the system if it is not in use.and another issue:The versions of ssh(OpenSSH) software used in the prompt are between 7.4-9.2, and there is a username enumeration gap corresponding to these versions. Various modified exploits of this vulnerability can affect up to ssh version 9.1.

I’ve been trying to get a CE up and running for a few days now.  I keep running into this issue:CVM will not start after initial install:[root@NTNX-0618718a-A ~]# virsh start NTNX-0618718a-A-CVMerror: Failed to start domain 'NTNX-0618718a-A-CVM'error: Hook script execution failed: internal error: Child process (LC_ALL=C    PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin /etc/libvirt/hooks/qemu NTNX-0618718a-A-CVM prepare begin -) unexpected exit status 1: [Errno 2] No such file or directory: '/sys/bus/pci/devices/10000:01:00.0/driver_override'Hopefully someone can point me in the right direction.  I’ve not been able to find a solution searching the forums.  I pulled the NVME out of the mix and I”m only using ssd’s and spinners for data. I can ssh into ahv just fine but CVM shows:[root@NTNX-0618718a-A ~]# virsh list --all Id   Name                  State-------------------------------------- -    NTNX-0618718a-A-CVM   shut offSystem Hardware:Fujitsu Primergy RX2530 M4 256GB RAM d

Is there anyway to collect multiple VM’s mac address and vLAN & IP details through prism central.

I installed CE and can’t install a compatible version of prism central. How can I get access to prism central?  

Hello Experts,It is possible to update the cluster to a version of AOS that supports ESXi 6.7 U3. However, we do not have a license for ESXi 7 or 8.We must upgrade and deploy the PC. The current version is 5.20.3.5. After reviewing the compatibility matrix on the Nutanix portal, I discovered that this version of AOS 6.5.3.7 is compatible with ESXI 6.7 U2 and U3. did any one tested that before ?and i would like to deploy PC,PS : this cluster has esxi 6.7 u3 as hypervisor  

Hello Community, hope you are all doing well.  We have 8 NX-3155G-G6 nodes and Each node has 4 x 8TB HDD and 2 x 3.84TB SSD.We would like to add new disks to expand to storage Capacity. After checking this link , we noticed that there are 6 empty drive slots on each node and it is not clearly specify how to fill them (please have a look on the picture below).What are the possible layouts to add new disk to populate the empty slots ? Thanks in advance.

Hello together,i have followed the installation guid:https://portal.nutanix.com/page/documents/details?targetId=Nutanix-Community-Edition-Getting-Started-v2_1:top-login-wc-ce-t.htmlnow the only thing left is to connect via the web console but the browser cant connect to it it refuses the connection due to timeout. Do i have to aktivate the web console first or is there any config that i could have missed?

Newbie trying to configure Nutanix.Setup:Dell server with iDRAC VLAN’s are setup on the switch Nutanix is installed. Both the Host and the CVM have an IP address in the same VLAN (100)Questions:Should I configure the VLAN on the CVM first or on the host? It seems if I configure the CVM first I can then configure the host. If I configure the host first I can’t get to the CVM. (expected) When I configure the CVM it never seems to return so I have to hard reset the host. Setting the CVM tag back to 0 it does come back.Commands:On the host: ovs-vsctl set port br0 tag=<vlan tag> On the CVM: change_cvm_vlan <vlan tag>

Hi all, new to Nutanix here likely for the same reason a million other folk are trying it out. I’ve managed to find the download (though not exactly straight forward) for Phoenix 6.8.1 CE and am attempting to first time install on an r820 with 5 enterprise ssd sas in front bay. I’ve split the first off into it’s own virtual drive at full capacity (300something gb) for H install. The second and third off into their own individual virtual drives for CVM1 and CVM2. Then the last two combined into one virtual drive for D.  I am able to select the virtual drives for H and D however the CVMS tell me it can’t be selected because the drives must be SSD which they are. Drives are fine and the virtual disks are recognized just not allowed to be selected. Any insight is greatly appreciated I’d love to test Nutanix out but documentation has been all over and even downloading (from a link in a seemingly random forum post) my confidence is a bit waned.

It’s been running for 21 hours. Is there a way to stop, delete so I can re-deploy? Thanks

Good day, I have deployed a 3 nodes so that I can play around with. all is fine until I’m about to mount the cluster:3 nodes on a same vLANAHV can be pinged from my laptop but cannot ping the CVMcan Ping each of the CVMs from their respective AHV Until nowmanage_ovs show_uplinks  2024-09-06 19:27:37,861Z WARNING manage_ovs:1263 Failed to fetch gflags. Acropolis service might be down: HTTPConnectionPool(host='127.0.0.1', port=2030): Max retries exceeded with url: /h/gflags?show=hypervisor_username (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7f688f73f590>: Failed to establish a new connection: [Errno 111] Connection refused',)).2024-09-06 19:27:37,864Z WARNING manage_ovs:1263 Failed to fetch gflags. Acropolis service might be down: HTTPConnectionPool(host='127.0.0.1', port=2030): Max retries exceeded with url: /h/gflags?show=hypervisor_ahv_nonroot_support (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTP

Hello,I am attempting to deploy a Nutanix CE virtual machine on KVM; however, I am encountering the following error:  Upon investigation, it appears that the installer is attempting to access certain system information from /sys/class/dmi/id within the VM, but this data is unavailable.I have made several attempts to pass this information using virt-install, but these efforts have not been successful. Does anyone have suggestions on how to successfully install a Nutanix VM on KVM? virt-install \--connect qemu:///system \--cpu=host \--virt-type=kvm \--hvm --accelerate \--arch x86_64 \--features acpi=on,apic=on,pae=on \--name nutanix1 --ram 49152 --vcpus=24 \--disk path=/root/vm_images/nutanix1-disk1.qcow2,size=60,bus=virtio,format=qcow2 \--disk path=/root/vm_images/nutanix1-disk2.qcow2,size=200,bus=virtio,format=qcow2 \--disk path=/root/vm_images/nutanix1-disk3.qcow2,size=200,bus=virtio,format=qcow2 \--cdrom /root/phoenix.x86_64-fnd_5.6.1_patch-aos_6.8.1_ga.iso \--boot cdrom,hd \--netwo

Hello guys,I am facing Index error while Nutanix CE setup installation in VMware workstation test LAB, Please suggest solution(index Error: list index out of range)Give below resources for Nutanix VM Memory 32Processor 8HDD 65 GB  

Hi TeamDoes Nutanix support IPV6 based VM deployments. Currently I am using Prism Central V3 API for deploying VM https://www.nutanix.dev/api_reference/apis/prism_v3.html#tag/vms/paths/~1vms/postIn the payload section it is mentioned that IPV4 support is only available now. If we want to do IPV6 based deployment, is there a workaround for that? Please share the document reference if IPV6 based VM deployment is not possible. If a workaround is there, please let me know how to do this. ThanksAkash

HiI am trying to install the community edition on three nodes and on each one it keeps on getting stuck on this screen I have tried multiple times and wait over 30 minutes any ideas? 

I have installed CE 2.0 on 3 Supermicro servers. AHV is up, and the CVM is running. The Genesis service is running on all 3 nodes. When I issue the command to create the cluster and use a comma separated list of IPs, I receive the following error (same error regardless of which CVM I run the command from):MainThread cluster:1001 Could not discover all nodes specified. Please make sure that the SVMs from which you wish to create the cluster are not already part of another cluster. Undiscovered ips : localhostThe localhost part is throwing me off. Any ideas? I can ping all the CVMs from all the other CVMs. I have restarted nodes, restarted genesis, and used the --skip_discovery with no luck. Any help would be appreciated. 

I’ve attempted to install a single node cluster (many times) and seem to have the same issues.  Post install, the cluster is not created.  Genesis appears to not be installed or running? I can ping and ssh between AHV and CVM.  Some issues I see leveraging the ‘troubleshooting common nutanix issues’: 

I’ve tried installing AHV/CVM and have ran into a multitude of issues.  However, I thought I finally got a successful CE up.  I was able to SSH to CVM, verify Genesis was running.  Yet I randomly get issues with this VM and the network and now, yet again, I can’t access the CVM or even the GW.  I’m following this doc: :https://www.jeroentielen.nl/installing-nutanix-community-edition-ce-on-vmware-workstation/.I’m also doing some troubleshooting from: https://portal.nutanix.com/page/documents/kbs/details?targetId=kA00e000000LM8BCAW and not getting anywhere.Sometimes I can ping the GW, sometimes I can SSH to the CVM, sometimes I can’t. 

Erasure coding is a data protection method that breaks data into pieces, adding redundancy for fault tolerance and data recovery. When we enabled RF2 on the cluster that copies data in two different nodes for fault tolerance, should I use erasure coding on the cluster? what is the benefit?

Hi TeamI am trying to execute some NCLI command to interact with my Cluster. I have download the nCLI artifacts from my Prism Central web UI. Followed the steps to install it on Windows machine as mentioned here https://portal.nutanix.com/page/documents/details?targetId=Command-Ref-AOS-v6_6:man-ncli-session-local-initiate-t.htmlBut when I am trying to connect, I am getting the below error.PS C:\Users\maharanaak> ncli -s 10.108.60.35 -u admin -p -jUser Name: adminEnter Password:com.google.gson.stream.MalformedJsonException: Expected EOF at line 1 column 5Cannot invoke "java.util.Map.get(Object)" because "messageBody" is null I have tried with both CVM IP and Prism Central IP and in both cases got the same error.Attaching the screenshots of installation related to java version, ncli version and the command execution here for reference.Please help here.@Kcmount  @JeroenTielen  

Hi! I’m trying to complete my first Prism Central deployment on a test lab, however whenever I try to connect my cluster to the PC, I get the following error, any idea on where I may have went wrong? I can login to the PC just fine, and everything pings each other with no issue.