Search results

Ah, thank you. The "place in templates/cache folder" was the bit I was missing! (I think - will need to experiment) Thank you for responding.

Any update on this? I just tried pct set 149 --template 1 and got implement me: template at /usr/share/perl5/PVE/LXC/Config.pm line 1027. Which I guess is the same issue being discussed above.? I guess I can accept a "do it this way instead", but ultimately I got misled by the documentation...

An ssh approach does work. It would be nice if the pct documentation included something like "pct exec has restrictions compared to, say, an ssh approach. Notably, the execution environment in the container ceases to exist when pct exits."

I wish to launch a script on a collection of containers. pct exec seems to do what I want. Except I want to background my script and have the pct exec finish immediately. I tried just adding an & (ampersand) but when the pct exec finishes, it terminates the backgrounded script. I have now tried...

I have a 16 core/32 thread ThreadRipper (AMD 1950X). root@zen:/var/log# numactl --show policy: default preferred node: current physcpubind: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 cpubind: 0 1 nodebind: 0 1 membind: 0 1 root@zen:/var/log# numactl...

Hi - I'm in the process of learning about NUMA on my ThreadRipper. Here's a blog summarising someone else's learning: https://yunmingzhang.wordpress.com/2015/07/22/numactl-notes-and-tutorialnumactl-localalloc-physcpubind04812162024283236404448525660646872768084889296100104108112116120124/ I...

Ah, ok. I ended up temporarily removing the mp. I thought I'd read something that a workaround - well, more of a workaround than just hiding the mount. But I must have been wrong. Thanks for resonding :)

I am running current (pve-manager/5.2-9/4b30e8f9 (running kernel: 4.15.18-5-pve)) pve and getting an error during container cloning. My container has a mount point, and I remember reading something about mounpoints can be awkward with clones but can't find it now. The conf file for the container...

I have real temperatures and clock frequencies now. Although temps are not per core, it lets me check my confidence in my cooling setup, which is the key thing I wanted to verify. Thank you again for the guidance.

Ok, thanks for the reply. I shall go and have a go at building this myself and see what happens.

The it87 driver is something I saw referenced when trying to understand what wasn't working, so it's interesting you reference that. It's clear I need to be running a "suitably" recent version to have Ryzen family recognised. But, lacking much familiarity with all of the conventions behind how...

I have pve with a 4.15 kernel. I am trying to get lm-sensors to tell me the individual core temperatures for the CPU. It looks like TR family (17h) is not recognised, either by sensors-detect or some other part of my configuration. I had hoped this would "just work" with 4.15, but obviously...

I now have the swap setup I was trying to achieve - thank you. Migration probably isn't an issue for me (yet) as this is just my home server. But I will bear that in mind if it ever becomes relevant.

Ah, thank you! I shall try this when I get home this evening, but it makes sense.

I am running threadripper with X399 Taichi mobo. I have ECC ram and ZFS striped across 2 discs. Nice little home setup. The 4.15 kernel, which will be released soon, brings per-core temperature monitoring, which you might like. I am running headless as a server though - I've seen various posts...

My proxmox host has an SSD that I want to use as swap space for one of my VMs. I don't understand how to connect it though. If I try to add another disc as hardware in the VM hardware tab, it only lists my ZFS stuff - but the partition I want to use is just a raw partion (/dev/nvme0n1p6)...