Deployment Success

Have a customer that we just installed new 6035 nodes using the Haswell chip. Found that they had an issue where CVM network latency checks would periodically fail. Inspecting the log with support showed occasional 70ms+ spikes both for CVM-to-CVM and CVM-to-gateway settings. This is a vSphere 5.5 U2 cluster, so we set the power settings for high-performance instead of balanced (default by foundation) and the latency issue went away. We're guessing there is some new ACPI tweaks required by VMware for ESXi. After configuring high performance, the latency between CVMs went from .2 to .3 millisecond average (typical) to more in the .07 - .09 range, and so far we're no longer observing any big spikes.

Hi, There is a fun bug with version 4.1.2.1 ... (or its me?.. :cathappy: ) In version 4.1.1.4, it was possible to update Hypervisor via PRISM UI (with version 5.5u2), but in version 4.1.2.1, its not possible :(see below) Thks guys ! Sig'

Hi Everyone, I've planed to deploy Spunk on Nutanix on my Company. But i'm newbie in Nutanix, so I have some question about this: 1. Nutanix have hypervisor which I can use is ESXi, HyperV,KVM... so in my case, I will use Esxi and rehat to install Splunk. it's mean I have install Splunk on Redhat which is running on ESXi, is it right? Can i install Redhat directly to Nutanix server (1 node have 4 server?) 2. About the Cluster, Which is control the cluster on VM Host? Vmware vCenter or Nutanix Prism Central? 3. Nutanix have an OS is VM Controller? is it install on Vmware machine or is it a hypervisor? Thanks for all reply to make my clear this stuff.

So the hosts themselves have been changed accordingly for the timezone the customer wants, but was wondering, for Prism to show the correct timestamps on stuff being done, would this come from the CVM time zone or where is the setting for the switching PRISM to use the customer specified timezone for timestamps?

I'm trying to find out what the default networking setup looks like for a hyper-v cluster. If we were to take a 3x3060 for example, I assume the 10GB ports are teamed and then both added to a external switch of similar? what about the 1gbs ports? if anyone has a couple of screenshots that would help my curiosity... Cheers Dave

Hello, I'm tring to reset my IPMI password using ipmicfg-win.exe for WIndows, but I got the following error: Can not find a valid IPMI device. Help me! Thanks...

Hi Team , I would like to know what is the default cluster id on a new nutanix node in the config json file. I have edit my json file and unable to create cluster where failed genesis error. I would like to know the default phrase? example : {"rackable_unit_serial": "14SM15290032", "node_uuid": "f771faf1-5cc4-4653-961f-cad2fa96a15f", "node_serial": "ZM144S054999", "node_position": "A", "cluster_id":"15311", "rackable_unit_model": "NX-1050"} What cluster id i should use? tried delete it but still cant create cluster.

Hi, I am new to Nutanix world. I would like to know about the MSL 2024 FC Tape drive solution integration with Nutanix Servers. I have found the Network FC Cards for Network cards but not able to find FC HBA for external storage connectivity. Please reply. Thanks.

Hi Guys, Ive been trying to reimage a block of mine but it keeps failing. from the foundation logs , it keep showing this message. im able to access to IPMI directly using the ip address. Stopping imaging thread poolImaging thread pool stoppedTraceback (most recent call last): File "/home/nutanix/foundation/lib/installer.py", line 390, in service_imaging_thread foundation_port) File "/home/nutanix/foundation/lib/installer.py", line 365, in do_imaging imaging_step.pre_threading_prep() File "/home/nutanix/foundation/lib/imaging_step_type_detection.py", line 187, in pre_threading_prep raise StandardError(error_message(result, node_config.ipmi_ip))StandardError: Foundation couldn't reach the IPMI interface at 192.168.100.101. Please double-check your IP addresses and network setup.

Any thoughts about moving the CVM to a non-daylight saving timezone We typically use the GMT timezone of Atlantic/Reykjavik.

Hi guys, i'm new to nutanix and was assigned to take nutanix. I would need some helps in further understanding all about it. Any rules, foundation, rules, knowledge i should know ? What are the more common technical problem? Any help would be appreciated.

I can't really find good instructions for this or else I am just not understanding, but if we are adding new nodes to an existing Nutanix AND Hyper-V cluster and therefore are not creating a new storage cluster name and what not, can the setup_hosts script be used in that case? I was thinking from looking at that, it does more than just rename hosts, add to domain, tell it to use 10Gb and reboot. Essentially, I have new nodes with Hyper-V and NOS installed and i know I could easily expand the Nutanix cluster itself from Prism, but for the piece of adding the new hosts to the domain and sucking up into SCVMM, i wasn't so sure. I was initially thinking i would just run the "setup_hosts" script and skip the "setup_scvmm" script (as i am using existing failover cluster from the first blocks installed), however looking at the setup_hosts script, i see it prompts for actually configuring the public address of the SMB share and/or name, etc. I don't really need or want to create a ne

Hello folks, We recently installed NOS version 4.1.1.4 on all our clusters and Prism Central. Some folks have asked if we can increase the timeout for the Prism Central. How can this be done? Thanks.

Remote Site Configuration: IP Address and Port Input Field Hello everyone Can I use the Port-Filed in the Remote Site Configuration? What's the default Port? Is there a port range nutanix is listening for Remote Site Replication?

Does cluster stop command have impact to VMs ? Will VMs in the cluster operating normal?

Hi Everybody, I'm new to Nutanix, I'm having my first node installed on Monday so ery excited ! I have the chance to dedicate a Nutanix host for a Lab and will have the time to play around with it. I've seen that 192.168.5.0/24 is a dedicated subnet for the CVM, but this is also the subnet in which my Lab resides. Is it ok to use the CVM and the production VM in the same subnet ? What are the options ? I've been told the CVM network cannot be changed. But is it ok to have the VMs and the IPMI in the same LAN ? For a production environment I can understand that security wise and performance wise, this is mandatory to split the subnets but in my case this is a bit different. Thanks. Cheers. Fred.

With the release of cloud connect, you can backup to Amazon using a CVM running as an instance. Is the CVM cloud agnostic? Can we run it in other environments, such as vCloud Air, Azure or partner vClouds?

When we received email alert generated by CVM, the timestamp in the message follows PST time zone (which I believe it is CVM default). Can we configure CVM to use our local time zone?

Trying to find the best procedure for the following. Had 36 nodes. Imaged all with foundation and IP'd in a standalone environment with private IP's on a basic switch, in an effort to image them all with ESXi at one time. No datastores were created yet using that process. Had to split up those 36 imaged nodes to move to 4 separate networks to create 4 separate clusters (9 nodes each). EAch network is completely isolated from the others, and not connected the internet. Started with one block (4 nodes) -- moved to new network, manually re-IP'd everything for that one block to have valid Network1 IPs. Now I am wanting to move the other 2 blocks to this first network. Am I forced to do a manual re-IP of each node now, after having used foundation to IP them in the beginning with private IP's? Prism cannot see (discover) the new nodes (via IPv6) after connecting them up to the same network. And the re-IP IPv6 cluster_init page is not reachable, . Just wondering if there is a way to b

Hello - In Prism on an 8-node cluster, we are getting an alert under Health showing 3 of 8 CVMs that their Memory Usage are @ 100%, for about a week now as shown in the graph (in Prism). However, those CVMs seem fine in vCenter. At a high level, Guest Mem % for the CVMs are around 40% or less. Thoughts? Thanks.

Is there a metric I can add from the Analysis section of the Web UI that would show me stats on a Protection Domain replication?

When I go to edit the Break/Fix options for my products, I can't change their location - I can only change the support personnel. We're just getting up and running with our Nutanix hardware and we're moving one of our Blocks to a DR data center in a few days; our primary data center is where the Block has been assigned to.

One of our newly installed Nutanix blocks shows 0 VMs from the web interface. There are no user VMs on this block; wouldn't I at least see a controller VM per node (3 nodes currently so I think I should see 3 controllers VMs). The cluster is up and all shows as healthy. Running vSphere on these nodes and when I connnect to the nodes indiviually via vCenter, I can see the controller VM. What am I missing?

Nice little Guide to upgrade Esxi to version 5.5 http://joshsinclair.com/?p=278

Hi, in the platform admin guide is mentioned that there can be only 1 node failure at one time. What happens if a complete block get's lost? thanks and best regards