• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Issue After changing IP of a server, slave DNS manager stop to work

SalvadorS

Regular Pleskian
Server operating system version
Debian 11
Plesk version and microupdate number
18.0.49
Hello,

After having changed the IP of one of my plesk servers, the slave DNS manager tool which is installed in this server has stopped to work. The zones are not sync to the slave dns server.

I delete the slave servers configuration in the slave DNS manager tool and added them again but still the same problem.

So I was wondering in uninstall the "slave DNS manager" tool and reinstall it again. But it is impossible to uninstall from the control panel.

So, anybody have any suggestion to fix the problem or how can I uninstall slave DNS manager from a SSH access?
 
That's an extension. Why can it not be uninstalled form the Extensions page? What is the error message?
 
Hi Peter,

Thanks a lot for trying to help me. I don´t know why, there is only a "open" button, not an uninstall one...Captura de pantalla 2023-02-24 a las 11.20.32.png
 
The problem with the extension is that it tries to transfer zones from the old IP instead the new one:

Feb 24 11:21:16 ns1 named[21159]: transfer of 'aDOMAIN.com/IN' from X.Y.Z.A#53: failed to connect: timed out

Where domain.com is the domain on the server and X.Y.Z.A the old IP that had the server, not the new one...
 
There should be two buttons to the right of "Abrir", I do not know why they are missing. Instead, on the console you can run
plesk bin extension -u slave-dns-manager to uninstall.
 
Thanks again Peter.

I uninstall the tool and reinstall it but I have the same problem


Feb 24 11:36:26 ns1 named[21159]: zone DOMAIN.COM/IN: Transfer started.
Feb 24 11:36:26 ns1 named[21159]: zone DOMAIN.COM/IN: got_transfer_quota: skipping zone transfer as master X.Y.Z.A#53 (source 0.0.0.0#0) is unreachable (cached)

Where domain.com is the domain on the server and X.Y.Z.A the old IP that had the server, not the new one...

Any ideas?
 
Hello Peter,

Yes, I followed that article to change the IP of my plesk server, and worked fine as always, but this is the first time I find this problem with the slave DNS manager tool
 
Also I checked that a second server with I changed IP exactly 7 days ago have the same problem. So It seems that the slave DNS manager tool is working with the old IP instead the new one. And uninstall and reinstall the tool does not fix the problem...
 
Hi everyone,

Hmm... If I not mistaken, the "Slave DNS Manager" extension does not have a listener for server's ip changing events. I would suggest to re-add a secondary DNS server to the extension. It also makes sense to check the configuration on the secondary DNS server.

  • A tooltip for the "status" icon contains info about what command was executed and what status was received during the execution. An argument for the "-b" flag should be an IP-address on the server that should be used to communicate with the secondary DNS server. If there is everything is ok but some error still happened, make sense to check configuration on the secondary server.
1677309166329.png

  • The extension suggest adding the next configuration on the secondary server (see an example below). There is also used an IP-address of the server to Plesk to control who can change records on the secondary DNS server. It make sense to check the configuration and update it to using new IP-address of Plesk server.
Code:
/*
    Add to named.conf on the remote server (config location depends on the OS of the slave server):
   
    options {
        ...
        allow-new-zones yes;
    };
   
    key "rndc-key-192.0.2.1" {
      algorithm hmac-md5;
      secret "blablablablablablablablabla==";
    };
   
    controls {
        inet * port 953 allow { 192.0.2.1; 127.0.0.1; } keys { "rndc-key-192.0.2.1"; };
    };
*/


I believe there are all required steps to make everything working again :)
It also make sense to us to think about improving the extension to update server's ip automatically but it still will be required updating configuration on the secondary DNS server.
 
Hallo,

thanks for this help - we had the same problem.

Instead of reinstalling/reconfiguring the Slave DNS manager it is sufficient to change the IP configuration in the database.

So our steps were:

1) Remove all synced zones on the slave server - commands can be generated with a "one-liner":

Bash:
named-nzd2nzf /var/cache/bind/_default.nzd | grep OLDIP | awk '{ print $2; }'|cut -d'"' -f2 | awk '{ print "rndc delzone -clean " $1;}'

Execute the generated commands on the slave server.

2) Change IP address of primary server in the Plesk "psa" database:

SQL:
# get module settings for DNS manager / old IP
select * from ModuleSettings where value like '%OLDIP%';

# update OLDIP for DNS manager - you have to replace the values with the result from the first query
update ModuleSettings set value='NEWIP' where module_id='XX' and name ='masterIp-SLAVEIP';

3) Sync zones :)

Everything was fine after those steps.

We did it in this way, because we had several hundreds of zones - so this was a quick&easy way.

Maybe it might help others to solve the problem.

Best regards,
Jonas
 
Back
Top