Initially, there should be two groups: account and domain admin. Apparently they are.
[root@nethserver ~]# /usr/libexec/nethserver/list-groups -s
{“account”:{“members”:[]},“domain admins”:{“members”:[]}}
OK, so listing of users doesn’t work. AD is ok, groups are ok.
Do you see the groups “account” and “domain admins” in the server manager?
Do you have usernames including special characters?
I don’t see groups or users in the server manager.
There is only one user and it does not contain special characters: teszt.
Please check following…
Check members of groups:
/usr/libexec/nethserver/list-group-members account
Check groups of user:
/usr/libexec/nethserver/list-user-membership teszt
It seems to me it is right:
[root@nethserver ~]# /usr/libexec/nethserver/list-user-membership teszt
[“account@mydomain.eu”]
[root@nethserver ~]# /usr/libexec/nethserver/list-group-members account
[“teszt@mydomain.eu”]
I think the problem is that the users can’t be listed so they’re not shown in server manager but I don’t know why.
Maybe rejoining AD from command line helps, here is a similar problem (old thread):
Do you mean that? I’ve already read …
Isn’t this a single command? Or two?
find /var/lib/sss/ -name ‘*.ldb’ -delete > /etc/sssd/sssd.conf
Shouldn’t you sleep? I don’t want to hold you up slowly I should sleep too. I can’t stay awake anymore … I’ll have a long day tomorrow … Will we continue tomorrow? In the meantime, thank you for your help.
1 Like
Two. The second one empties sssd.conf.
You’re welcome. OK, let’s continue tomorrow.
OK, I understand.
Thank you, continue tomorrow. Good night.
1 Like
Hello @mrmarkuz !
I shut down and restarted the new server several times yesterday to see if the error goes away. I had no luck, nothing changed. The same problem persisted.
Last night I no longer tried the fix you suggested and found in the forum, I turned off the new server and went to bed.
I recently got home, turned on the new server, and saw that users showed up! The cockpit does not write an error, but a new error appears in the old Server Manager:
"Enable admin user
Enable and set a password for the admin user. "
I attach a picture of it.
I really don’t understand anything. I don’t know what’s going on with the server and why. I have no idea why I can’t restore the config one time. I don’t know what preparations are needed to restore the config and make the new server work.
Can you edit the users in cockpit?
These are not really errors, just configuration warnings.
Cockpit introduces new backup methods so maybe old server manager shows it as a warning, I think you can ignore it.
Did you set a password for the admin user before the backup?
Hello, I’m home again. It was a long day… I’m reading your answer now.
Yes, I can edit users.
Previously, before the config backup, I set a new password for the admin on the source server to create my own databases in phpmysqladmin.
Recently, I temporarily set a new password for the admin on the destination server in the cockpit to resolve the error message. After that, I restored the data backup made on the source server to the destination server because I knew that it would restore the mysql databases and so the admin password would also be restored. The data restore was running correctly. I didn’t receive any error messages.
I have just started to check the operation and correctness of the restored services and data. I haven’t found any errors yet, the VPN, databases are available but I’m not done yet.
After that, I start all over again because I never manage to do a correct recovery. But before that, I should know what I’ve done wrong so far so I don’t repeat mistakes.
This may be followed by an attempt to install the working server on a VPS because it was still not successful and is very important due to the installation of CentOS minimal and the dummy interface.
How can we determine why the config failed to restored again? What could be the reason that the config failed to restored again? It would be very important to know this …
You may look at the history to check which commands you executed between the poweroff and the reboot.
We need to find errors in logs or by executing the commands directly.
Unfortunately, a data restore restores the source server’s bash history to the destination server.
I did not execute any commands yesterday before shutting down the server, which allowed users to recover after turning on today.
I might be able to search the messages log, but I don’t see any recovery entries in it.
Do you have any idea?
No, I’m out of ideas but the restore worked, didn’t it?
It seems there were issues resolved by a reboot/poweroff.
Unfortunately, this did not happen. After restoring the config, I restarted the server several times yesterday and the users never recovered. It’s unclear what happened on today’s lunch break when I powered on it.
Did you try if the AD survives another reboot/poweroff?
I’m afraid we won’t find errors when it works.
To check the restore process I recommend using a VM and snapshots. This way you can easily rollback to see if you can reproduce the issue and the way to solve it.
1 Like
I turned off the server last night and turned it on this morning and everything seems to be working, the AD survived turning it on and off.
I fear that the cause of the problems will not be revealed and will recur every time.
A snapshot is a good idea if provided by your VPS provider, but I am now using a local non-virtualized server. That’s why I created a new VMWare virtualized server to test the recovery, but it also fails to restore either the config or the data backup from the working server. Recovering a VPS from backing up a non-virtualized local server is another task because we got stuck with it before.
You suggested that I try to restore the configuration and backup to another server with a non-virtualized physical network interface (not a dummy interface). This experiment is happening now, but it wasn’t really successful either.
Again, I’ll start step-by-step through a new installation and recovery to see the results of applying a few changes. It would be good to create a reassuring safe procedure (scenario) …
Sorry but if everything is working now, what do you fear?
Are we talking about the same dummy interface explained here?
I was talking about the dummy interface created in the guest OS, a virtual interface in virtualization layer (VMWare) is no problem.
Contabo for example provides it.
The basic problem is installing the working server on a VPS by restoring the backup on the VPS. That failed. Therefore, for testing purposes, I tried to restore the backup of the working server to another physical server to filter out these virtualization issues. However, it didn’t work perfectly either, but for unknown reasons it works …
Yes, I’m talking about this dummy interface solution. This is described as need to install below CentOS Minimal before installing Nethserver. Otherwise, Nethserver will not recognize and use the dummy interface as a Green LAN interface. Nethserver requires two network interfaces (Red and Green) to install …
Due to many problems and other database problems unfortunately it is too early to look for a VPS provider, but thank you for the suggestion.
1 Like