Hello,
First time installed for CPU folding, not received WU yet. However I have a question if William updates the Client how should we proceed to update ova? we dont have to re-deploy right?

Hi all,

(in the v1.01 appliance) Did anyone attempting to install the NVIDIA drivers get this error?:

"ERROR: Unable to find the kernel source tree for the currently running kernel. Please make sure you have installed the kernel source files for your kernel and that they are properly configured; on Red Hat Linux systems, for example, be sure you have the 'kernel-source' or 'kernel-devel' RPM installed. If you know the correct kernel source files are installed, you may specify the kernel source path with the '--kernel-source-path' command line option."
(after running ./NVIDIA-Linux-x86_64-440.64.run as listed on page 2, step 4)

I can kind of get around it by using the ' --kernel-source-path="/usr/src/linux-headers-4.19.112-1.ph3/" ' parameter but should there be a need to?
(I say kind of as nvidia-smi finds no devices, but that could be due to other issues).

You shouldn't have to do this with NVIDIA GRID, Tesla, or Quadro drivers. What kind of card are you using, and how are you attaching it to the VM?

I'm seeing if I can get an RTX 2060 going via PCI passthrough, so not a supported config. Do you know if the GRID/Tesla/Quadro drivers are precompiled?

The GRID/Tesla/Quadro drivers are not precompiled. However, the docs have directions for compiling the drivers.

Since you're using a consumer grade card, you may run into other issues.

Did you run this line from the GPU instructional PDF?

tdnf install -y libstdc++ gcc linux-devel make binutils glibc-devel

Yup! Though as a consequence of running it (before the attempt to install drivers) it did upgrade the kernel package to 4.19.112-1.ph3 from 4.19.104-3.ph3...

]# tdnf install -y libstdc++ gcc linux-devel make binutils glibc-devel
Package libstdc++ is already installed.

Installing:
mpc x86_64 1.1.0-2.ph3 photon-updates 363.64k 372372
libgomp x86_64 7.3.0-4.ph3 photon 192.77k 197400
binutils-libs x86_64 2.32-3.ph3 photon-updates 2.64M 2764136
libstdc++-devel x86_64 7.3.0-4.ph3 photon 13.12M 13756707
libgomp-devel x86_64 7.3.0-4.ph3 photon 331.07k 339011
libgcc-devel x86_64 7.3.0-4.ph3 photon 132.00b 132
libgcc-atomic x86_64 7.3.0-4.ph3 photon 32.21k 32984
glibc-devel x86_64 2.28-4.ph3 photon-updates 10.88M 11403890
binutils x86_64 2.32-3.ph3 photon-updates 24.89M 26103467
make x86_64 4.2.1-3.ph3 photon-updates 1.10M 1158101
linux-devel x86_64 4.19.112-1.ph3 photon-updates 41.60M 43624632
gcc x86_64 7.3.0-4.ph3 photon 113.53M 119041682

Total installed size: 208.66M 218794514

Upgrading:
linux x86_64 4.19.112-1.ph3 photon-updates 25.82M 27074405

Total installed size: 25.82M 27074405

Downloading:
mpc 114195 100%
libgomp 100711 100%
binutils-libs 645284 100%
libstdc++-devel 2208475 100%
libgomp-devel 108979 100%
libgcc-devel 9220 100%
libgcc-atomic 21250 100%
glibc-devel 3061763 100%
binutils 8130151 100%
make 449070 100%
linux-devel 12854374 100%
gcc 42586282 100%
linux 22142239 100%
Testing transaction
Running transaction
Installing/Updating: linux-4.19.112-1.ph3.x86_64
initrd generation of kernel 4.19.112-1.ph3 will be triggered later
Installing/Updating: libgcc-atomic-7.3.0-4.ph3.x86_64
Installing/Updating: libgcc-devel-7.3.0-4.ph3.x86_64
Installing/Updating: libstdc++-devel-7.3.0-4.ph3.x86_64
Installing/Updating: binutils-libs-2.32-3.ph3.x86_64
Installing/Updating: libgomp-7.3.0-4.ph3.x86_64
Installing/Updating: libgomp-devel-7.3.0-4.ph3.x86_64
Installing/Updating: mpc-1.1.0-2.ph3.x86_64
Installing/Updating: gcc-7.3.0-4.ph3.x86_64
Installing/Updating: binutils-2.32-3.ph3.x86_64
Installing/Updating: linux-devel-4.19.112-1.ph3.x86_64
Installing/Updating: make-4.2.1-3.ph3.x86_64
Installing/Updating: glibc-devel-2.28-4.ph3.x86_64
initrd of kernel 4.19.104-3.ph3 removed
Removing: linux-4.19.104-3.ph3.x86_64
(re)generate initramfs for 4.19.112-1.ph3, transfilertriggerin 2.0-5.ph3
Creating /boot/initrd.img-4.19.112-1.ph3

Complete!

Looks like things are finally working (assuming this first work unit uploads successfully). Thanks all!

For anyone who might be attempting the same thing, here's what worked with and what did/didn't make a difference (in this specific instance):

CPU: E3-1246 v3
Motherboard: ASRock H97M Pro4
BIOS: 2.10
GPU: GeForce RTX 2060
ESXi: 6.5.0, build 13635690
NVIDIA Driver: 440.64 (NVIDIA-Linux-x86_64-440.64.run)

What worked:
- ASRock BIOS
--- Primary graphics adapter had to be set to to PCI instead of Onboard in BIOS
- Passthrough enabled devices assigned to VM (not sure if all were required)
--- 0000:01:00.0 Available nVidia Corporation VGA compatible controller
--- 0000:01:00.1 Available nVidia Corporation Audio device
--- 0000:01:00.2 Available nVidia Corporation USB controller
--- 0000:01:00.3 Available nVidia Corporation Serial bus controller
- VM memory
--- 6 GiB (All reserved. Went to 6 instead fo 4 on a whim)
- VM advanced configuration parameters
--- hypervisor.cpuid.v0 set to FALSE

What was tried but apparently wasn't needed (in this specific instance):
- Upgrade of VM hw version (v11 works)
- Changing the VM firmware boot option from 'BIOS' to 'EFI' (works when set to 'BIOS')
- Adding the VM advanced configuration parameters pciPassthru.use64bitMMIO=TRUE and pciPassthru.64bitMMIOSizeGB=<n> (this gpu only requested a mapping of 16 MiB + 256 MiB + 32 MiB = 304 MiB)
- Modifying '/etc/vmware/passthru.map' (the default setting of "10de ffff bridge false" worked in this instance)

Thanks all for your work on this project/appliance!

New note for those having difficulty SSH'ing to your VM...

I was apparently having a problem with special characters in my password. In this case, an ampersand. ("&")

I deployed a new instance and changed the password and I can now SSH to the machine without issue.

Is anyone else having a problem connecting the FAHControl app to the FaH client on the VM? I've tried from two different machines (Win10 and MacOS) with the same result. FAHControl shows the client as "Online" like my other machines. But when you click on it to view the system, the "Status" column on the left changes from "Online" to "Updating" and it won't actually populate any information on the client's operations.

I've tried both with the system's hostname and by IP.

The client was not configured with a password and I did not change any of the subnet information. (It's all either 0.0.0.0/0 or 127.0.0.1 - as default.)

This is a direct LAN IP connection. There are no firewalls or security measures in between.

Felt the need to add that, yes, the VM is running. The FAH client is operational as shown on the console in Workstation/Fusion/Remote Console. It's crunching away.

Alright. Disregard. It appears that I both overlooked a crucial detail in the "Deployment Steps" PDF and then, also, misinterpreted it. (I'm multitasking quite a bit here.)

For those who might be in the same place as I and also didn't correctly read the documentation:

"vii. F@H Remote Mgmt password
Note: This sets a password to be used when connecting FAHControl to your Appliance. The default is VMware1!"

What confounds me though is that I specifically left the password field blank when I was deploying the OVA to vCenter. So that's a little ambiguous.

But I'm now connected and showing stats as normal. ;)