Search results

Manufacturers do switch it up sometimes, but on this one my bullshit meter is off the charts. comparing the back of the link to a genuine card, the difference is extremely apparent. Intel E810-XXVDA4 E810XXVDA4LG1P5 Quad Port 25GbE PCI-E 4.0 NIC Server – Voyage Technologies vs...

This does not look like a genuine Intel adapter to me. The heatsink is not the same as the original heatsink for this card, and it covers the area where the GPIO pins for external PPS input for PTP timestamping should be, I also see no brady intel sticker for verification on the back of the...

@NaCl Then I'm sorry to say I'm drawing a blank, might be a built in safety against cross flashing by intel. I've only had to do this for firmware rollbacks

@NaCl Hmm. might be some specific code that cant be overridden in config, but if you want to yolo this, i would try this with the first party intel updater: ; NIC device BEGIN DEVICE DEVICENAME: E810_CQDA2_O_SEC_FW VENDOR: 8086 DEVICE: 1592 SUBVENDOR: 8086 SUBDEVICE: 000B NVM IMAGE...

@NaCl on the topic of forcing an upgrade though, this is the process: in the folder you extracted the firmware updater to, find the nvmupdate.cfg eththool shows you the firmware version, in my case this is: firmware-version: 4.20 0x80017785 1.3346.0 so in the nvmupdate.cfg there is a...

@NaCl Are you completely sure these are genuine cards? or possibly some OEM variant? The naming looks completely off to me. My E810-CQDA2 appears like this on both hosts with E810-CQDA2 adapters. Intel(R) Ethernet NVM Update Tool NVMUpdate version 1.40.5.5 Copyright(C) 2013 - 2023 Intel...

I've had a mellanox connectx-3 40gbe adapter in my DS 1821+, they work fine, and runs the same driver as the 10GBE variant, though I tested it on version 6.2, I've not tested it on DSM 7, since the connectx3 was repurposed for something else, and it's now running an intel X710-DA2. Though MLX4...

I replaced ESXi with KubeVirt installed on RKE2, Ubuntu OS, with Rook Ceph storage. Looking at replacing Ceph with 2 standalone active/passive failover NFS nodes, since Ceph performance is absymal even on NVME storage with 5 nodes with appropriately scaled PGs on rados block device pool on...

Does cockpit-storaged support configuring an iscsi target? I thought it was primarily for managing the linux host as an initiator connecting to iscsi targets?

If you have 10 or fewer physical servers to back up, the free Veeam community edition might work? https://www.veeam.com/veeam_data_platform_feature_comparison_ds.pdf Free Backup Software For Windows, VMware, & More - Veeam

try pbd-unplug xe command reference | XCP-ng Documentation

Yes. xe should be available from your default shell on the server.

I see. It should be noted that I've not used xenserver since it was open source distributed from citrix itself without limitation, so it's been a few years, but as I recall, you should be able to detach and remove the existing SR. If it's not possible by GUI, then your best bet is: (edit i'm...

I see. You only have 1 drive then? XCP-ng is opinionated in that it assumes you have a dedicated drive for the hypervisor's "privileged os" or "dom0", so it consumes the entire drive for that purpose with the file system of choice by the developers of xcp-ng for stability. My suggestion here...

You could use the command line for xe sr-create with type=ext4 if you want to force it to ext4.

Based on those critera, I would agree that XCP-ng is the more appropriate choice, since it has a free LTS release built for stability, while with Proxmox, the stable enterprise repo is paywalled and only the upstream repo is available for free. If you wanted a more bleeding edge homelab, I...

No. No. No. Let me stop you there. KVM is not a slow hypervisor and it's not bogged down by huge overhead. Amazon Web Services literally replaced their Xen based hypervisor (which is what XCP-NG is based on) in favor of KVM. AWS adopts home-brewed KVM as new hypervisor • The Register Edit...

In that case, there is not really any distinction between ESXi, XCP-NG, etc. KVM is still a type 1 hypervisor using hardware extensions for virtualization, IOMMU, etc, as I said, qemu just manages the type 1 KVM hypervisor from userland when it runs KVM workloads. I'm not aware of any problems...

@Octopuss Type 1 and Type 2 hypervisor doesn't really apply that well to QEMU/KVM. KVM is a type 1 hypervisor built into the linux kernel, QEMU is a type 2 hypervisor in user-land, but the way this is done is managing KVM virtual machines through KVM extensions built into QEMU, because thats...

I had another look at the datasheets for SN2100 and SN2700, you are correct. I assume these are running the same switchchip, and that the SN2100 doesn't need to disable ports by virtue of having fewer ports to begin with and not needing to switch lanes between ports. sn2100 sn2700