[BlueOnyx:07253] Re: DNS Script ?

[Ken - Precision Web Hosting, Inc]

----- Original Message ----- 
From: Dan Porter
To: BlueOnyx General Mailing List
Sent: Saturday, May 07, 2011 10:15 AM
Subject: [BlueOnyx:07251] Re: DNS Script ?


LOL, its not difficult.   The hosting customer does not have access to the 
second box and it's one of those petty little maintenance tasks that could 
be automated and eliminated.  I thought perhaps someone on this list had 
already done so but maybe I am the first to consider it.   We have also had 
the case of someone mistyping the domain name on the secondary DNS service.

Thanks Jeff !!     Dan.




---------- Original Message ----------- 
From: Jeff Folk <jfolk at qzoneinc.com>
To: BlueOnyx General Mailing List <blueonyx at mail.blueonyx.it>
Sent: Sat, 7 May 2011 11:36:18 -0500
Subject: [BlueOnyx:07250] Re: DNS Script ?

> On May 7, 2011, at 11:06 AM, Dan Porter wrote:
> >
> > So the question was, is there such a script available that someone has 
> > already created to automate the creation of the secondary records based 
> > on those from the primary - OR - what are the ramifications of simply 
> > running two primary DNS servers by just FTP'ing over the zone files when 
> > a change is detected on the first primary DNS server?
>
> What is so difficult about adding a domain name and IP into the secondary 
> dns service on the second box? It isn't like you have to recreate all the 
> dns records... The complete zone file is transferred to the secondary dns 
> server (as long as you have given permission to the IP address in the 
> primary server).
>
> _______________________________________________


Dan

What I do is on the server hosting the sites and the dns is :
set up a cron
22  10  *  *  * /root/domainlist.sh

That script contains:
#!/bin/sh

rm /home/sites/somesite.onyourserver.com/web/secondaryrecords.txt
ls -1 /home/sites | grep www. >> 
/home/sites/somesite.onyourserver.com/web/secondaryrecords.txt
chmod 766 /home/sites/somesite.onyourserver.com/web/secondaryrecords.txt
perl -p -i -e 's/www\./secondary - /g' 
/home/sites/somesite.onyourserver.com/web/secondaryrecords.txt
perl -p -i -e 's/\n/ 123.456.78.90 - \n/g' 
/home/sites/somesite.onyourserver.com/web/secondaryrecords.txt


That file contains records like the ones below, where 123.456.78.90  is the 
IP of your primary nameserver.
secondary - oneofyourdomains.com 123.456.78.90 -
secondary - anotherofyourdomains.com 123.456.78.90 -


Then on the secondary server, I get those records and inject them into the 
named.conf

35 21,7 * * * /root/getrecs.pl


That file contains:

#!/usr/bin/perl

$aproblem = "nothing";

system ("rm secondaryrecords.txt secondaryrecords.txt.* ");


$filetoget = "http://123.456.78.90/possiblysomedir/secondaryrecords.txt";
$wgetinfo = `wget  $filetoget 2>&1`;
if ($wgetinfo !~ /200 OK/i) { print "We have a problem with $filetoget , 
will exit. Info is $wgetinfo";  exit; }


$filetoget = "http://123.456.78.99/possiblysomedir/secondaryrecords.txt";
$wgetinfo = `wget  $filetoget 2>&1`;
if ($wgetinfo !~ /200 OK/i) { print "We have a problem with $filetoget , 
will exit. Info is $wgetinfo";  exit; }

$filetoget = "http://223.456.99.101/secondaryrecords.txt";
$wgetinfo = `wget  $filetoget 2>&1`;
if ($wgetinfo !~ /200 OK/i) { print "We have a problem with $filetoget , 
will exit. Info is $wgetinfo";  exit; }


system ("cat secondaryrecords.txt secondaryrecords.txt.* > allrecords");

open (IN, "<allrecords")  or die "Couldn't open input CSV file: $!";

my @allsortedrecords = sort <IN>;

system (" cp /root/orignamed.conf /root/named.conf");
open(NAMEDCONF," >> /root/named.conf");

$lastline = '';

foreach $thisline (@allsortedrecords) {
   ($prev0, $nothing, $domain, $ipaddress, $prev4) = split(/\s+/,$thisline);
   #print "prev0 $prev0, nothing $nothing, domain $domain, ip $ipaddress, 
nada $prev4";
   #print "The domain is $domain \n";
   if  ($domain eq $lastdomain) {
        print "$domain eq $lastdomain ----------------------------- \n\n";
        $aproblem = $lastline;
        #exit;
        next;
   } else {
      if (($domain =~ /\s/) or ($ipaddress !~ /\d+\.\d+\.\d+\.\d+/)  or 
($domain =~ /in-addr.arpa/) or  ($thisline eq
        $lastline) ) {
         print "domain $domain  then ip is $ipaddress or thisline eq last 
$thisline eq $lastline next\n";
         next;
       }
      print NAMEDCONF "zone \"$domain\" \{\n";
      print NAMEDCONF "  type slave;\n";
      print NAMEDCONF "  file \"db.$domain\";\n";
      print NAMEDCONF "  masters \{ $ipaddress; \};\n";
      print NAMEDCONF "\};\n\n";

     #if ($lastline eq "precisionweb.net") {
     #  print "This line is $thisline";
     #  print "Last line is $lastline";
     #  exit;
     #}
      $lastdomain = $domain;
   }

}
close(NAMEDCONF);


system ('rm -f /var/named/chroot/etc/named.conf');
system ('cp   /root/named.conf  /var/named/chroot/etc/named.conf');
#system ('/etc/rc.d/init.d/named restart');

$restart =`/etc/rc.d/init.d/named restart`;

print "The restart is $restart";


if ($restart =~ /FAILED/) {
print "We have a problem with restart. It was $restart \n\n";
#copy the old file in and try again
#email me

} else {
print "All is fine\n The restart was $restart";

}



if ($aproblem ne "$aproblem" ) {
   print "A problem line was $aproblem\n";
}




So, it gets secondary lists from multiple servers, sorts them, checks for 
duplicates, injects them into a template file, then copies that to the 
named.conf  I'm sure there are better ways to do it, but that is what I use 
for customer dedicated VPS


----
Ken M
Precision Web Hosting, Inc.
http://www.precisionweb.net