John Jawed (JJ)

> Maybe there's a better way to access all available GPUs that is maintained by nvidia and work with all the possible env variables at once. I’ll go investigate this...

Still working through this. I don’t want to exec out to nvidia-smi for this. So far the alternatives I’ve explored to that are: 1) emulating nvidia-smi - traverse /dev/nvidia* devices...

And one more … 3) use https://github.com/Cldfire/nvml-wrapper

Those screenshots help, can you also please ls /etc/yum.repos.d/ and if possible provide the contents of CentOS-Base.repo ?

Can you run the script again?

That helps even more 😃 I think the issue is that epel is providing libidn2-2.3.0-1.el7.x86_64 ... which stops the whole install. Can you please disable epel and try again? We...

I haven’t fiddled with 7.9, let me take a look.

Not quite - definitely an area which requires some attention. I also need to get the updated sources from internal repos over to here.

There's integration with foreman's puppetrun API if you want to use that...otherwise foreman integration is not necessary. On a puppet managed host you would deploy the pvc::agent - there's a...

There is a cron entry created which runs every minute as part of the pvc::ppm manifest, it collects load + passenger statistics and sends them to the PVC server. These...