Hi, I recently setup a second node and came across a few things that might be useful to add to the Toolbox

Third up, is enabling IOMMU. This differs for BIOS (Grub) and UEFI (systemd-boot).

Enable IOMMU For BIOS/GRUB

Reference:
https://pve.proxmox.com/wiki/PCI(e)_Passthrough

Follow the linked instructions.

For the kernel command line, add:

GRUB_CMDLINE_LINUX_DEFAULT="${GRUB_CMDLINE_LINUX_DEFAULT} intel_iommu=on"

To a new file:

nano /etc/default/grub.d/iommu.cfg

Save, exit, and update grub

update-grub

Enable IOMMU For UEFI/Systemd-boot

Reference:
https://pve.proxmox.com/wiki/Host_Bootloader#sysboot_edit_kernel_cmdline
https://matthewdeporter.com/proxmox/proxmox-pci-passthrough/#iommu-systemd-boot

Add the ''intel_iommu=on'' ''iommu=pt'' settings to the Systemd-boot command line.

Edit the /etc/kernel/cmdline:

nano /etc/kernel/cmdline

Before modification:

root=ZFS=rpool/ROOT/pve-1 boot=zfs

After:

root=ZFS=rpool/ROOT/pve-1 boot=zfs intel_iommu=on iommu=pt

Update the Systemd-boot scripts:

pve-efiboot-tool refresh

Verify loaded kernel cmdline:

$ cat /proc/cmdline
initrd=\EFI\proxmox\5.4.73-1-pve\initrd.img-5.4.73-1-pve root=ZFS=rpool/ROOT/pve-1 boot=zfs intel_iommu=on iommu=pt

Verify that IOMMU is enabled and Virtualization Technology for Directed I/O:

$ dmesg | grep -e DMAR -e IOMMU -e AMD-Vi
[ 0.013784] ACPI: DMAR 0x000000007E2CAB18 000130 (v01 A M I OEMDMAR 00000001 INTL 00000001)
[ 0.155343] DMAR: IOMMU enabled

[ 1.474961] DMAR: Intel(R) Virtualization Technology for Directed I/O


Hello,

thank you so much for this great script.

Could you please add an backup-and-restore option where PBS and PVE is installed on one server.

That would be great.
Thx

Hi, I recently setup a second node and came across a few things that might be useful to add to the Toolbox

First up, is terminal colours! I'm not a fan of how stock Debian keeps everything the same, but I am a fan of how Turnkey Linux has theirs setup.

This can be done by editing bash.rc

nano /etc/bash.bashrc

Before: From Line 18

# set a fancy prompt (non-color, overwrite the one in /etc/profile)
# but only if not SUDOing and have SUDO_PS1 set; then assume smart user.
if ! [ -n "${SUDO_USER}" -a -n "${SUDO_PS1}" ]; then
  PS1='${debian_chroot:+($debian_chroot)}\u@\h:\w\$ '
fi

After:

# set a fancy prompt (non-color, overwrite the one in /etc/profile)
# but only if not SUDOing and have SUDO_PS1 set; then assume smart user.
if ! [ -n "${SUDO_USER}" -a -n "${SUDO_PS1}" ]; then
    eval "`dircolors -b`"
    alias ls='ls --color=auto'
    alias ll='ls --color=auto -alF'
    alias la='ls --color=auto -A'
    alias l='ls --color=auto -CF'
    alias grep='grep --color=auto'
    alias fgrep='fgrep --color=auto'
    alias egrep='egrep --color=auto'

    # Set a terminal prompt style (default is fancy color prompt)
    PS1='${debian_chroot:+($debian_chroot)}\[\033[01;33m\]\u@\h \[\033[01;34m\]\w\[\033[00m\]\$ '
else
    alias ls="ls -F"
    alias ll='ls -alF'
    alias la='ls -A'
    alias l='ls -CF'
    PS1='${debian_chroot:+($debian_chroot)}\u@\h \w\$ '
fi

Rebonjour
Suite à l'erreur (voir autre issue) sur PBE, je teste sur PVE.
Voici l'erreur que j'ai et que j'ai du mal à interpreter.
Je n'ai pas d'IPv6 configfuré sur mon serveur PVE (juste une IPv4 fixe locale)
IPv6 est tout de même activé sur mon réseau. PBE semble en ipv6 juste avoir son IP de loopback en fe80

`- Checking for known errors that may be found in logs
Nov 4 09:35:15 pve postfix/smtp[209517]: connect to mail.infomaniak.com[2001:1600:0:aaaa::1:2]:465: Network is unreachable
Nov 4 09:36:10 pve postfix/smtp[209792]: connect to mail.infomaniak.com[2001:1600:0:aaaa::1:1]:465: Network is unreachable
Nov 4 09:36:10 pve postfix/smtp[209792]: connect to mail.infomaniak.com[2001:1600:0:aaaa::1:2]:465: Network is unreachable

• Are you on IPv4 AND your host can resolve and access public adresses? y = yes / anything = no: ^C
  `

Comment puis-je en arriver là ?
J'ai indiqué yes dans la question mais cela ne semble pas améliorer la situation.
Pas sur que ça soit une issue rééelle, plutôt conf, mais dans le doute je poste quand même

EDIT :
Et la config du postfix généré par le script

`myhostname=XXXXX.XXXXXXXx

smtpd_banner = $myhostname ESMTP $mail_name (Debian/GNU)
biff = no

appending .domain is the MUA's job.

append_dot_mydomain = no

Uncomment the next line to generate "delayed mail" warnings

#delay_warning_time = 4h

alias_maps = hash:/etc/aliases
alias_database = hash:/etc/aliases
mydestination = $myhostname, localhost.$mydomain, localhost
relayhost =[mail.infomaniak.com]:465
mynetworks = 127.0.0.0/8
inet_interfaces = loopback-only
recipient_delimiter = +

compatibility_level = 2

smtp_use_tls = yes
smtp_sasl_password_maps = hash:/etc/postfix/sasl_passwd
smtp_tls_CAfile = /etc/ssl/certs/ca-certificates.crt
smtp_sasl_security_options = noanonymous
smtp_sasl_auth_enable = yes
sender_canonical_maps = hash:/etc/postfix/canonical
inet_protocols = ipv4
`

Hello,
Pour commencer, un immense merci pour ton script qui facilite bien la vie.
J’ai constaté un petit problème, lorsque l’on change la valeur de swappiness à zéro par exemple, c’est effectif tant que proxmox ne redémarre pas. Après un reboot, la valeur revient à celle par défaut: 60.
J’ai testé ça sur plusieurs hosts avec la dernière version à ce jour de Proxmox, la 7.0.-13

Hello, due to the grub/zfs non efi bug encountered now with pm 8.1 upgrade I have to reinstall one node after the other.
Does the restore process work on a member node?

Thank you.

Bonjour, merci pour ce script fort utile et bien réalisé :)
J'ai tout de même un problème avec la configuration des notifications emails voici le log :

Jul 26 15:30:01 pve postfix/smtp[4074964]: 92D543404C5: to=[email protected], relay=mail.s.fr[xx.xx.xx.x]:25, delay=219299, delays=219295/0.04/2.9/0.94, dsn=4.1.8, status=deferred (host mail.s.fr[xx.xx.xx.x] said: 450 4.1.8 [email protected]: Sender address rejected: Domain not found (in reply to RCPT TO command))

Je précise que j'auto-héberge mon serveur de messagerie en passant par un relais smtp (mailjet, port 465)

Merci par avance pour votre retour
Sylvain

Hey,
Thanks for the script!

Today i have a question about the SMTP Module.
I have configured it, but the emails always arrive as root <[email protected]>.
Is there a way to change the root part to something else?

I allready tried to edit the canonical file and make a postmap but that didn't work as expected.

“If you reveive this, it means your email configurations looks correct. Yay!”

reveive -> receive

It's a minor thing but couldn't help to notice. Love your work with this tool. It was so easy to set up email notifications :)

Bonjour Tonton Jo
Voici le log d'erreur suite utilisation du toolbox sur un PBS lors d'un test d'envoi.
PBE fraichement installé et mis à jour avec ta toolbox en 2.0.12

- What is the recipient email address? :
[email protected]
- An email will be sent to: [email protected]
proxmox_toolbox/proxmox_toolbox.sh: line 554: mail: command not found
- Email should have been sent - If none received, you may want to check for errors in menu 3

Je n'ai aucune erreur dans le menu 3

`- Checking for known errors that may be found in logs

• No configured error found - nothing to do!

`

hi @Tontonjo

apt update
apt install shellcheck
shellcheck proxmox_toolbox.sh

see what happen

regards,
nGadmini binti nGadimin

Hi
Is it planed to have update from PVE 7.x to 8.x ?

Hello,

what argument do I have to use to Backup Configuration?

Thank you,

Bonjour TontonJo,
Merci pour ton travail que je suis depuis quelques mois.
Je viens de mettre à jour mon PM en V7 sans soucis grâce à ton tuto et je viens aussi d'utiliser ton toolbox.
Par contre je viens de faire une bêtise avec que je ne comprends pas et je ne m'en sors plus.
J'avais déjà un compte root sans autorisation en ssh et un compte que j'appelerai ici perso@pam autorisé en ssh.
j'ai lancé l'option 4) security pour activer fail2ban et j'ai dû ensuite faire une erreur dans les réponses suivantes (je ne voulais pas créer d'autres comptes) car je n'ai plus d'accès ssh avec mon compte "perso" (ni root bien sûr).
je peux toujours me connecter à la GUI (en root et en perso) et à l'intérieur à la console >_shell de proxmox pour rentrer des commandes.
ça fait 4h que je cherche sur internet comment réautoriser mon compte perso@pam en ssh et je n'y arrive pas. Pourrais-tu m'aider ?
Merci d'avance.
Fabrice

Hi, I recently setup a second node and came across a few things that might be useful to add to the Toolbox

Second up, is additional packages.

• lshw for identifying hardware
• lm-sensors for additional sensor information for fan control scripts etc
• corosync-qdevice needed for having a third 'qdevice' node in even numbered clusters. Often a SBC.

I'll add others as I think of them.

As the the title says the no-subscription message doesn't get removed after running update host. Tried running it a couple of times and also rebooted the server but I still get the no-subscription message.

Hi, I recently setup a second node and came across a few things that might be useful to add to the Toolbox

Third up, is CPU governor.

I generally change this to ondemand on a new install which reduces power consumption but doesn't seem to affect performance.

I do this with the below command:

# echo "ondemand" | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor

Then add a cron job to run at each boot:

crontab -e

@reboot echo "ondemand" | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor >/dev/null 2>&1

we had multiple crashed services after backup restoration.

We did a reinstallation and wont use the backup feature again.
Just wanted to let you know that it is currently unusable.