Search results

Thanks for that. The host was originally installed with PVE6. I don't recall the exact history, but I remember installing grub-efi-amd64 following the upgrade to PVE7. That process removed grub-pc. IIRC if I then purged grub-pc and reran update-grub, I ended up with a non bootable system. I'm...

The PVE host is setup for EFI only booting and I am using grub, not systemd boot. Some time ago, I switched form `grub-pc` to `grub-efi-amd64`. As you can see, grub-pc was installed, then removed and its configuration remains. So I was wondering: Do I still need `grub-pc-bin`? Can I purge...

Since I am booting via EFI, do I still need to retain grub-pc-bin package? # dpkg --list | grep grub ii grub-common 2.06-13+pmx2 amd64 GRand Unified Bootloader (common files) ii grub-efi-amd64 2.06-13+pmx2...

There doesn't seem to be anything amiss there. As you're aiming to 'pass through' to an LXC, any IOMMU config is not needed at this point. I have a much older intel cpu/igpu (vaapi). I can pass the dri device node through to multiple unprivileged lxc and they then successfully use the renderer...

Does your motherboard BIOS have any options that affect initializing video when both igpu and a discrete graphics card are installed?

I have recently noticed some unexpected changes to device nodes. A USB bus now gets assigned to /dev/bus/usb/002 where as it had been 001 previously. The intel gpu node /dev/dri/card0 has become card1 (now there's no card0). This seems to have occurred since kernel 6.8. The USB numbering, I...

If you do think the internal received headers influence google's treatment of the messages, try setting the PMG mail proxy option: 'Hide internal hosts'.

I avoided running docker on the pve host. and went with a VM, despite the fact I do not have the PVE firewall enabled. Though you can have pre/post exec systemd overrides to handle the mangling of the firewall. Then I wanted separate stacks, on each of 3 subnets. I could not resist the...

If you want to geo block with iptables you should see better performance with ipsets. I am not sure ufw does that for you. IIRC I used that approach with geoblocking on a consumer grade router years ago and it handled it ok.

Firstly I haven't done this myself but AIUI postfix re-attempts delivery at specified intervals (default every 4 hours and gives up after 24 hours). This is configurable, directly in postfix if not by PMG. The PMG web UI has a deferred mail report which gives you a view of queued mail. As to how...

Many thanks for the clarification. I'd assumed postfix ran chroot. The comment in master.cf is misleading as it implies the postfix default is to run chrooted. # ========================================================================== # service type private unpriv chroot wakeup maxproc...

I notice postfix check complains about differences between system files and their counterparts in postfix chroot: warning: /var/spool/postfix/etc/localtime and /etc/localtime differ warning: /var/spool/postfix/lib/x86_64-linux-gnu/libnss_systemd.so.2 and...

I believe the match is case insensitive, unless it's a regex. A colon is not required.

And it was a particularly inopportune moment to make that mistake with the 8.2 release just appearing.

Whilst a BIOS update might address those ACPI errors, you mention it's been running fine shortly before that. So, unless you also updated the BIOS before the issue arose (or updated the kernel), I think it unlikely those error messages are connected to the problem. Many/Most motherboards display...

This may have been caused by unavailable DNS. If I access the PVE web UI by IP, rather than hostname, I am able to get much quicker VNC console access to the pfsense guest and so see the splash screen and get to the guest's boot menu. I expected the desktop OS or web browser to have cached DNS...

I find, when trying to reach the console of a vm when booting it, vnc is slow to connect. Meaning I miss the Proxmox splash screen/bios and the opportunity to interact with a guest's boot menu. e.g. pfSense and trying to interrupt boot to get single user mode. Is there something I can do to...

If the CT's OS is recent debian, use journalctl command. Same with the PVE host. What I would do is tail the journal journalctl -f on the PVE host, then start the CT. Perhaps do the same in the CT guest, tail the journal prior to starting docker and any containers. journalctl will, if terminal...

As @leesteken says, you may be confusing nested kvm with the nesting feature of CTs. The latter I believe allows the CT access to the host's procfs and sysfs. Something I think you need in order to run docker in an unprivileged CT. I have a couple of CTs successfully running docker and PVE is up...

Ah, I misunderstood your question. I (mis)read it as your trying to use the hosting company as smarthost for PMG.