Service configure not updated

[Headmaster]

Hi All,

I had encounter a problem on my new RedHat 7.2 Server, whatever I updated the dns record, add a sub domain or ever enable ssh access for one of the user though DirectAdmin, the daemon didn't reload the config file until I manually restart it.

I had tested on the following case :

1. Add / remove a dns record though DirectAdmin, the named daemon up time looks updated ( I guess is using kill -HUP ?), the config files (/etc/named , /var/named/??? ) looks good, but actually the dns can't reply the resolve request until I manually restart the service.

2. Add a new domain or adding a new user though DirectAdmin,
the apache still directing the web page to default page of the server ip, not the virtualhost setting in the httpd.conf, need to restart the service to activate the change from DirectAdmin.

3. Grant a SSH access privllage to the user, chechked the log that showing the /etc/passwd had been modify to right shell, but can't use the SSH access( secure log show "not allowed because not listed in AllowUsers" ) , restart the SSH daemon to make it work.


I am wonder is there any problem that my DirectAdmin dunno how to let the daemon reload the config files ? It seems so stupid that I need to keep updating the config files though a cron job.

Thx for help !

[DirectAdmin Support]

Hello,

As root from an ssh prompt type:service named reloadThis must work for the files to be reloaded. Some versions of rehdat come with a buggy versions of bind, but we have the instructions to fix it on our installer page, 3/4 the way down:
http://www.directadmin.com/installguide.html

John

[Headmaster]

Here is the /etc/rndc.conf

------------------------------------------------------------
options {
default-server localhost;
default-key rndckey;
};

server localhost {
key rndckey;
};

key rndckey {
algorithm hmac-md5;
secret "!!!!";
};

------------------------------------------------------------

I had enable the trace log of named and monitor the following log at the same time i add the sub-domain

tail -f /usr/local/directadmin/data/task.queue
tail -f /var/log/messages

After I add a new sub-domain, task.queue will show a new entry : action=named&value=reload

But the messages log don't get any named reload message, I had try to manually type "rndc -s 127.0.0.1 reload", and the log should show


Sep 7 03:24:36 localhost named[3354]: loading configuration from '/etc/named.conf'
Sep 7 03:24:36 localhost named[3354]: no IPv6 interfaces found
Sep 7 03:24:36 localhost named[3354]: zone new.sub-domain.com/IN: loaded serial 2003090705
Sep 7 03:24:36 localhost named[3354]: zone new.sub-domain.com/IN: sending notifies (serial 2003090705)

I guess there is some problem the directadmin can't pass the command to rndc ?

[DirectAdmin Support]

Hello,

All DirectAdmin will do for a reload of named, is "service named reload". It assumes that all the proper reload information is the /etc/rc.d/init.d/named boot script, and that it's working properly. On top of that, I can't say much as we don't install bind, it's supposed to be on the system before DirectAdmin is installed.

You might have a look at the named boot script to see what the "reload" command will do.. and any customizations you want to have it do during a reload can be added.

Joh

[westm003]

i have the very same problem...

Reloading from the comment line is working like a charm...

But nothing is reloaded by directadmin!!!!!!!!! :mad:

I manually reload after a change and i run a cronjob once an hour.... :confused:

I need a fix...

Centos 4.4

[DirectAdmin Support]

Try:wget -O /etc/init.d/named http://www.directadmin.com/namedJohn

[westm003]

i replaced the binary (/usr/local/directadmin/) dataskq... from a working server....

and it started working directly!!!!


:cool:

[cobrax]

DirectAdmin Support DirectAdmin Support is offline
Administrator

Join Date: Feb 2003
Posts: 5,082
Try:
Code:

wget -O /etc/init.d/named http://www.directadmin.com/named

John
Reply With Quote

this now working for me after days for searching