Proxmox VE Configuration

All docs related to Proxmox VE configuration

[How-To] Add Web Server to Proxmox VE Node

Introduction

This allows you to access Proxmox VE via the port 443

Tested from Proxmox 3.4 - 6.3

Why do I need this?

Sometimes there is a firewall restriction that blocks port 8006 and since we shouldn't touch the port config in proxmox we'll just use nginx as proxy to provide the web interface available on default https port 443. Now let's begin...

Configuration

Install nginx

apt install nginx -y

Remove the default config file

In newer PVE and Debian versions:

rm /etc/nginx/sites-enabled/default

Create a new config file

nano /etc/nginx/conf.d/proxmox.conf

Note: You can choose the configuration filename freely, but it must have a .conf ending.

The following is an example config that works for the web interface and also the noVNC console:

upstream proxmox {
    server "YOUR.FQDN.HOSTNAME.HERE";
}
 
server {
    listen 80 default_server;
    rewrite ^(.*) https://$host$1 permanent;
}
 
server {
    listen 443;
    server_name _;
    ssl on;
    ssl_certificate /etc/pve/local/pve-ssl.pem;
    ssl_certificate_key /etc/pve/local/pve-ssl.key;
    proxy_redirect off;
    location / {
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade"; 
        proxy_pass https://localhost:8006;
	proxy_buffering off;
	client_max_body_size 0;
	proxy_connect_timeout  3600s;
        proxy_read_timeout  3600s;
        proxy_send_timeout  3600s;
        send_timeout  3600s;
    }
}

When doing this for a proxmox backup server, user the dir /etc/proxmox-backup/proxy.pem and /etc/proxmox-backup/proxy.key as that is where they are located. Also, use port 8007 as that is the port that pbs listens on.

Change the FQDN part to the fully qualified domain name of your host, you can check cat /etc/hosts output to find yours. in my case it was pve-dev-machine.proxmox.com. save the file and then check the syntax:

nginx -t

you should see:

nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful

If you see this message then everything should work and you can proceed.

systemctl restart nginx

After nginx service restarts you should be able to reach the web interface via either https://your.fqdn.goes.here or https://your.ip.address.goes.here

Post Setup

As the certificates reside on /etc/pve which is provided by the pve-cluster.service we need to tell nginx.service to only start after that one.

The easiest and cleanest way to do that is to add a Requires and After as a systemd override snippet.

This can be done with systemd edit UNIT which opens your $EDITOR:

systemctl edit nginx.service

here add:

[Unit]
Requires=pve-cluster.service
After=pve-cluster.service

and save + exit.

Enjoy the web interface on HTTPS port 443!

[How-To] Join PVE Node to Active Directory Domain

Purpose

This tutorial will walk through the steps necessary to join your Proxmox VE server to Microsoft Active Directory on a Windows Server.

Prerequisites

List of prerequisites:

Standard Name

You will be able to sync your users and groups into the Proxmox Permissions Panel in the GUI, and log into the web console with active directory credentials.

Here are the general steps:

  1. Set basic settings
  2. Add Active Directory Realm & Sync Users
  3. Configure Permissions
  4. Install packages from the repository
  5. Join the Active Directory domain
  6. Test your config


In this tutorial, our realm/domain will be infraredhat.com
Two domain controllers have been configured, and are acting as DNS servers.
Our subnet is 192.168.11.0/24

The Proxmox node a single network interface with a static IP of 192.168.11.10/24
The Domain Controllers (and DNS servers) have IP's of 192.168.11.11 and 192.168.11.12.

Set the search domain to infraredhat.com, and set the DNS servers to the IP addresses of the Domain Controllers, 192.168.11.11 and 192.168.11.12.

Configure hosts file with your Proxmox server's FQDN (and hostname).

Navigate to the "Time" panel below "Hosts" currently selected in the menu and verify the correct time and Time Zone.

Add Active Directory Server

Next, Select "Datacenter" or the name of your cluster, and Navigate to Permissions > Realms > Add Realm > Active Directory Server.
Populate your domain and domain controller IP's like the example below.

Click on the "Sync Options" Tab.
You'll need Active Directory credentials to access domain controller users and groups. You can simply use the Administrator Account, but for more security, you can create a user account with read-only access to the domains objects instead. For demonstration, let's use the built in "Administrator".

For the "Bind User", you'll need to enter it a very specific way.
Navigate to your domain controller and open a powershell window as admin.

SUPER IMPORTANT!!! (AND NOT VERY WELL DOCUMENTED!!!)

For the correct string to enter for the Adminstrator in the infraredhat.com domain, you would enter the following command:

Code:
dsquery user dc=infraredhat,dc=com -name Administrator

This is the output, and you can copy and paste it directly into the field, "Bind User"

CN=Administrator,CN=Users,DC=infraredhat,DC=com

Enter the password for this user. For now you can ignore the remainder of the fields. I've set my default sync preferences.

Click OK to close and save.
You can now select your Active Directly Realm from the list (listed alongside PAM and PVE). Click SYNC from the menu and you should see your AD Users and Groups populate.

Configure permissions to tell Proxmox what access to give each user/group

Next, Navigate to the "Permissions" Menu > Add > Group Permission
I've selected the Administrators Group from Active directory, and assigned the PVE Administrator Role to this Group.
This way, any user in the AD Administrator group will also be a PVE Administrator.
Select '/' to give full access as well.
Click OK.

Install additional packages needed for system security

Next, Navigate to a shell window on your PVE Node. There are a few packages to install and configure.
Update your packages, and install adcli, packagekit, samba-common-bin realmd
The remainder of the required packages will be auto-installed and configured.

Code:

apt update
apt dist-upgrade

# install the following packages (use apt list to see if they are needed)
apt install adcli packagekit samba-common-bin

# install realmd
apt install realmd


Join the node to the domain

Next, test connection with the domain controller, and then join the domain. For additional options, see the man pages by running the command man realm. Since we're doing a high level walkthrough, I'm keeping it simple. Because no user is specified in the join command, realmd defaults to "Administrator" for this action.

Code:

# test realmd
realm -v discover infraredhat.com


You'll see an output with information about your DC and domain, along with additional required packages. Installing these in advance will cause the configurations to fail. Simply enter this command, substituting your own domain.

Code:

#join the domain
realm -v join infraredhat.com


Follow the prompts, enter the Active Directory Admin password when prompted and allow the sssd and additional packages install.
You are now joined to the domain and you should see your Proxmox node appear as a computer in Active Directory Users and Computers. Congrats!

Configure additional settings and test your config

You can edit the config in /etc/sssd/sssd.conf. Run

Code:

pam-auth-update

and select the appropriate option if you would like home directories created for AD users. There is more information available on the internet about sssd.conf. Test your configuration with the command

Code:

id administrator@infraredhat.com

Great post, thank you for this it really helped.
If anyone needs a way to filter only specific users from specific groups, this is how I got it to work.

  1. When configuring the sync options for Active Directory, under "user filter" use these search parameters:

    Code:

    (&(objectclass=user)(samaccountname=*)(MemberOf=CN=group_name_here,OU=name,DC=domain,DC=tld))

  2. If DOMAIN.COM, is my root Active Directory tree, and USERS is a subfolder, which contains my users, you need to add a security group inside of the subfolder. In this example I will call my security group "TECHS", this is a group inside the "USERS" folder under the DOMAIN.COM tree.

  3. Make sure your users are inside this "USERS" folder along with the "TECHS" group. Add your users to the "TECHS" group. Done.

  4. Now in proxmox your query should be as follows (using the example parameters above):

    Code:

    (&(objectclass=user)(samaccountname=*)(MemberOf=CN=TECHS,OU=USERS,DC=DOMAIN,DC=COM))

  5. Done. Now inside of proxmox datacenter view, if you go to "Users" you should see your Active Directory users, that were part of whatever group you added them too.

  6. Select "Permissions" and click on "Add" above, click "Add user permissions" give them whatever permission they need to have.

When they login, they just need to use their username, at least thats how it worked for me. If you add their full mail address, e.g. "user@domain.com" proxmox appends the domain anyways, so itll try to login as "user@domain.com@domain.com", hence why you just need to put your username, e.g. "user" and thats it.

[How-To] Unlock CT Manually via Shell

Purpose

Manually unlock CT via the proxmox node shell.

Prerequisites

List of prerequisites:

Unlocking the LXC container with the pct tool

Proxmox VE provides the pct tool to manage Linux containers (LXC). To manually have the lock removed, simply issue the command pct unlock 107. This removes the lock on container 107. Deleting the failed snapshot can be performed by issuing the command pct delsnapshot 107 vzdump.

Please refer to man pct for more detailed instructions.

[How-To] Delete CT Manually in Shell

Purpose

Manually delete CT via the proxmox node shell.

Prerequisites

List of prerequisites:

Deleting the LXC container with the pct tool

Proxmox VE provides the pct command to assist with container actions. In this case, we will use this:

pct destroy 107

 This command where the number is the CT number assigned to it. Please refer to this:

man pct

For more detailed instructions.

If you get an error with this check the error output with this: 

cat /etc/pve/lxc/107.conf

If the LXC is locked, it will not work. Refer to the unlock page for this.