RSAT The procedure number is out of range

Hi, I come from Indonesia, and I have been using nethserver 7.9 for 4 years.
and this morning when I opened RSAT Tools from Windows 10 to manage users, I got an error:

then some computers can’t join to the domain:
administrator user has been entered.

For the past 4 years I have always managed 800 AddC users using RSAT tools on my Win 10 and this is the first time a problem has occurred, luckily sysvol is accessible.
get someone to help me.

Do you have a snapshot/backup? Would be good before trying things…
In some cases it helps to just restore a configuration backup.
If nothing else helps you can restore from backup.

Are there errors in the logs?

Which version of Windows 10 are you using? (Check with winver.exe)

Let’s check the config:

config show nsdc
config show sssd

Is the time/date set correctly on client and server?

Can the client resolve the server and AD correctly from DNS?

nslookup ad.domain.tld.

Does it help to restart the NSDC service?

systemctl restart nsdc

Maybe it helps to upgrade the DC?

signal-event nethserver-dc-upgrade


I apologize because after this there have been too many complaints at my company.
my nethserver restarted and the problem was resolved.
It seems that if the nethserver is not restarted for 2 months it will experience slowness, high RAM, and refreshing the user addc will take a long time.

Restart whole NethServer is a way to identify if the issue is there. But… it’s not the solution, it’s a snooze button.

How much is the ram for NethServer, currently?

yes, I call it the magic button.
My nethserver is running in proxmox with the ram I gave 22GB for nethserver, and proxmox 128GB ram.
If the nethserver is not restarted for 4 months, it will use up 15GB of RAM.

The longer the nethserver is not restarted, the more performance decreases.

I don’t know why this happens, but when my nethserver is on baremetal, it rarely happens.

1 Like


I’m using Proxmox for my around 30+ clients, all running NethServer 7.9 as AD in Proxmox.
All these Proxmox are licensed with the simplest “Cummonity Support”, but this entails me to access of the “Productive” Upgrades for Proxmox, which also entails about only a third of the Kernel Updates (These require a reboot in Proxmox due to VMs and running Containers, in PBS it does not require a reboot) and therefore much less reboots than a well updated Proxmox wiith the NO-SUBSCRIPTION Repo (Like I’m using for Home / LAB Installs of Proxmox).

And even then, it’s usually every 3-4 Months an update comes by, requiring a reboot of Proxmox (Thus also NethServer running there. Most of my clients are not running a HA Cluster!).

This suggests to me that you don’t update your Hosts very often! Bad for security and features, Proxmox adds a lot over time…

In other words: If you keep your hosts current (updating!), the issue would not appear!


My 2 cents


There are several things that make me unable to always reboot the server, especially if it is Proxmox, one of which is that the company where I work operates 7x24 hours non-stop, and users always work through shifts.
and I am very lazy to restart running services manually (there are some services that I deliberately run manually) :grin:
if I always restart the server, they will bother me every midnight.

Your suggestions can be my consideration from now on.

1 Like


Some of my clients, albeit a few, also operate 24*7. These have full HA clusters of Proxmox, meanng I can migrate any VMs to another Proxmox Hypervisor (all the time running!). Even then, I’ll once in a while reboot the NethServer…

All the others I update in the time from 22:00-01:00.

I do use Zabbix Monitoring for all clients, this helps me to “know” what’s ging on in each network, and to know problems before my clients call me… :slight_smile:

It’s also one of the reasons I have a small Home LAB with Proxmox, so I know when updates are available, and can test for any issues before doing my clients update…

My new Odroid H3 as Proxmox:

Uses max 60W power, no moving parts, gets only a little warm, and has 2 x 2.5 GB/S NICs, max 64 GB RAM and a 250 GB NVME (system in ZFS) and 2x1 TB SSDs in ZFS mirror for VMs.
Cost: About 350 USD…

It’s also very small:

Despite the looks, there’s NO ventilator inside. You CAN add one in, especially if using this as PBS with rotating hard-disks…

My 2 cents

1 Like

And… currently is not a bug, but a Syadmin problem? I mean… if bare metal works nice, and PVE slows down NS… Should not be a bug.