5 Essential Tips for Maximizing Your Experience at Nutanix .NEXT for Bloggers
Hi @LinKan , from our side NTLMv1 was also disabled and we don’t have any issues with API calls.Best thing to do to investigate is to make a wireshark record of your API calls and then filter the result ?
Hi, according to this link there’s no official support neither compatibility with less than Debian 9 (so kernel 4.9https://portal.nutanix.com/page/documents/compatibility-interoperability-matrix/guestos I really advise to follow this compatibility matrix to get your workloads working, hosting really hold Linux version may create vulnerabilities issues in your infrastructure and won’t be stable with some Nutanix software.
Hi Marc,I already done that in our infrastructure too, not seamless cause it never work at all but manually with the 3-click method, you won’t break anything when working on license. On the 3-click method you just have to go to Nutanix Portal Licenses, then remove any licenses on cluster, then do it again selecting clusters you need.Why doing this twice ? cause licensing workflow is a little bit bug (specially if you got many licenses products like Pro/Ultimate)
HiDidn’t find anything, the closest information we could retrieve is using this API call https://www.nutanix.dev/api_reference/apis/prism_v3.html#tag/vms/paths/~1vms~1list/post
I have newly installed HPE DX380 servers, total 3 NODEs with 10G/25G Dual port NIC ports (2 Pairs = Total 4 Ports). I have Mellanox Network switch which is configured with 1 to 12 port on 10 GB Network. After installing the AHV,AOS & creating the Cluster, when trying to create the VS, it is showing 3 Network Cards in 25G group & 1 in 10G. Therefore I am not able to create a BOND. My question is, How come only one NIC is showing 10G speed & rest of them with 25G. Logically if I have already configured my Network switch with 10G speed for all the required ports, In Prism also it should reflect as 10G even if that NIC is supported 10/25G speed. Does anyone have any solution for this. I am missing something? Can we change the NIC speed manually on each Node? Thanks in advance. are you sure that 4/4 NICs are able to support 10/25GB/s ? maybe you got ¼ different and that’s why you see it separatively ?On the other side, from a network perspective, I assume you manually force
HiYou should try to add if possible the Files Analytics features , cause it seems to say what you need about custom reporting :) here’s the doc : https://portal.nutanix.com/page/documents/details?targetId=File-Analytics-v3_2:File-Analytics-v3_2
hello. I need vitio 1.1.5 driver package. where i can download. I could not find from nutanix website. Please help me. Please search twice...really…There’s version 1.2.3 to 1.1.4 right here https://portal.nutanix.com/page/downloads?product=ahv&bit=VirtIO
Please mention OS you’re trying to configure.It seems you didn’t add properly virtIo drivers (or at least a bad version)
HiIt’s not as easy as you think. Here are the main steps to follow, but need a bit of work :Create an autounattend file to auto-configure what you need inside Windows (and test it manually) Use an API call or IaC tools like Packer/Ansible to create a template/create a VM and use the file you created.I spent a lot of days/weeks to build and test the autounattend part (because corporation has a lot of standards and settings to set) but since, it works really smoothly. It takes around 1-2min to deploy a VM and around 8-10min to custom it to our standards before delivering it . Also, if you need some API doc, within your Prism Central you got everything you need :https://your_prism_central:9440/api/nutanix/v3/api_explorer/index.html
HiI had the same issue last friday, I had to physically reboot the server.As the KB advised, you should at least upgrade the controller firmware (you can do it through iLo webpage manually) but I’ll advised to use LCM firmware on your cluster to use the HPE SPP to upgrade ALL your firmwares.
If you have Nutanix file servers deployed at both locations you could set up Smart DR replication between them, that results in an active/passive setup where the remote copy is read-only accessible. Does that satisfy your use case? Here is a useful video overview with more information. Thank you for this answer This is very useful information But we have another situation. We need it for Multi-Side Kubernets Cluster. We need Storage Solution for replication Storage for Kubernets Cluster. Which one solution could you recommend us ? I miss some details as Kubernetes cluster may be very wide in term of configuration.The solution that @smackeroo said should really fit your need. In my experience, we were able to migrate K8s data from one Nutanix infrastructure to another using replication between storage container. But depends on your K8s configuration, you’ll need to re-link your PVC as well.
@JoseNutanix you’re true.In my case I needed UEFI (and I add VirtIO drivers as well, forgot to mention it in my previous post) After double check in my Packer Windows Server 2019/2022 project, I’m doing the following : cd_files = ["./scripts/autounattend.xml", "./scripts/enable_winrm.ps1", "./scripts/sysprep.xml"] image_name = "tpl-win_server_2022_autounattend" image_description = "{{isotime `2006-01-02`}} - Packer Template Windows Server 2022 UEFI autounattended" shutdown_timeout = "2m" shutdown_command = "C:\\Windows\\system32\\sysprep\\sysprep.exe /unattend:E:\\sysprep.xml /quiet /generalize /oobe /shutdown"autounattend.xml is loaded for the Windows auto-install and configuration, then, at the first boot, it launches enable_winrm.ps1 which configure WinRM and Windows Firewall to let Packer send the shutdown command. About the shutdown command, i’m doing a single sysprep in mode “generalize oobe shutdown” in order to get a sysprep everytime a VM is deployed fro
In addition to @JoseNutanix here’s how I made the trick :Download official Debian ISO you need (in my case it was 11 but now 12) Create a preseed.cfg file that contains every instructions you need for your custom configuration Compile your official ISO with the preseed.cfg file Upload the custom ISO to Prism Central Then from a Packer file, simply write a file with hardware you need and link it to the custom iso you’ve just upload. Finally, launch your Packer project, VM will be created, OS will be automatically configured (thanks to preseed.cfg) and once it’s done, Packer will shutdown, delete and create an image from your VM.The Packer steps aren’t so difficult if you’re used to work with, but the preseed.cfg writing may be a tough part (spent many days/weeks to make all steps I wrote , working) Note: for Windows people, steps beyond are the same, but you need to write and use an autounattend.xml file instead of preseed.cfg as well.
As I wrote, Async/NearSync works on the same mecanism (as you can simply change from one to another in your configuraiton) so i’d say yes.Moreover, when I see tables like this I assume it’s ok I’ll tell you to ask Nutanix support to be 100% sure
Alright, so to answer to your original post, as nearsync is built on async mecanism, i’d say CG are fully supported. https://www.nutanix.com/blog/nutanix-dr-multi-site-recovery About configuration of CG in a nearsync environment : https://portal.nutanix.com/page/documents/details/?targetId=Prism-Element-Data-Protection-Guide-v5_16:wc-dr-near-sync-c.htmlon step 3 within Configuring NearSync Disaster Recovery
We have one PC instance in PRD site now. My question is: In order to use protection policy to replicate PRD data to DR site(We plan to use nearSync), do we need to create a new PC instance in DR site?*It’s not PC linked so you can configure it without creating a new PC instance.In your case you may configure a async/nearsync DR protection domain only for your Prism Central instance. It seems one PC could manage multiple PEs, so one PC instance is enough. But what’s the function for DR site PC? Is it a mechanism to sync data from PRD PC to DR PC?You right, 1 PC can manage multiple PEs.DR Site PC is a high availability feature for PC only. For example, if you got 1 PC that manage all PEs and you lost it, it will remain online/up on another site. It depends on your needs actually, if you can afford to loose or not your PC.
Hello !What do you mean by “CG” ?
Hi ! finally found a solution to do what I need in PowershellAll explanation was made in my ServerFault post (question + solution) below :https://serverfault.com/questions/1135179/update-json-payload-using-powershell-and-nutanix-api/1135270#1135270
Hey ! meanwhile i figured out on the same statement, and according to Nutanix official support all metadata + spec + api_version block should be send back in the PUT request.It’s working when adding categories when categories were null, but it’s kind of complicated when it’s about to keep existing categories, + add a new one (currently working on that )
Hi @JoseNutanix , thanks for these details !Just a quick question, i’m actually able to retrieve/select spec,api_version and metadata from the GET request (that i’ll use in another request in payload) Do you know how it’s possible to “update” some property of the payload, i.e how do you add/remove a category property ? As retrieved data type is PSCustomObject, i don’t know if there any EASY method to work with payload data in powershell Thanks :)
Hi !Even though i’m able to get all VM details through GET REST API Call, i’m actually not able to update only one property.How can you update specifical property you retrieved (ex: category, disk size etc...)? Thanks a lot :)
@Rubén Bernardo Calvillo Marín here’s the stepdownload an official ISO from Microsoft (in my case Windows Server 2022) download Nutanix VirtIO drivers from Support PortalFollow the steps in that article https://thunderysteak.github.io/windows-setup-driver-injection and don’t forget to add drivers in every indexes of boot.wim and install.wim Once you’re done , you can upload the new ISO on Prism Central and properly boot on, with UEFI settings enabled :)
Actually i found out that i have to add VirtIO drivers to Windows Server 2022 iso (to every index of boot.wim and install.wim). By doing that, everything went well and i can boot on the image easily without error.
Well here’s the solution, even though the error message wasn’t really clear to me, it seems that thestorage_class: block needed Prism Element credentials rather than Prism Central. According to french support team, it may change in a future release but at the moment, just let use PE credentials and everything runs smoothly.
Already have an account? Login
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.
