Problem with mass change of TTL and Serial Number

[mirchevideas]

Hi I'm a newbie in this forum but I see few topics for mass ttl and serial number changes so I try to do that for me but there is a problem.

What I do:
1. Log on the server as root via SSH
2. cp -a /var/named/ /var/named-backup
3. cd /var/named
4. perl -pi.bak -e "s/14400/60/g" *.db
5. perl -pi.bak -e "s/86400/60/g" *.db
6. sed -i s/'[0-9]\{10\}'/`date +%Y%m%d%H`/ /var/named/*.db
7. rndc reload
8. service named restart

After that I check a random .db file in /var/named . The file is for domain domain.bg and everything looks good.

; cPanel first:11.34.0.11 (update_time):2014031309 Cpanel::ZoneFile::VERSION:1.3 hostname:schost1.serverdomain.com latest:11.36.0.11
; Zone file for domain.bg
$TTL 60
domain.bg.	60	IN	SOA	ns1-70-82.superdnsserver.net.	manager.someotherdomain.com.	(
						2014031309 ;Serial Number
						60 ;refresh
						7200 ;retry
						3600000 ;expire
						60 ;minimum
	)
domain.bg.	60	IN	NS	ns1-70-82.superdnsserver.net.
domain.bg.	60	IN	NS	ns2-70-82.superdnsserver.net.
domain.bg.	60	IN	A	193.107.70.82
localhost	60	IN	A	127.0.0.1
www	60	IN	CNAME	domain.bg.
ftp	60	IN	A	193.107.70.82
cpanel	60	IN	A	193.107.70.82
whm	60	IN	A	193.107.70.82
webdisk	60	IN	A	193.107.70.82
webmail	60	IN	A	91.215.216.58
default._domainkey IN TXT "v=DKIM1; k=rsa; p=MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQDX7bbaE6IP1yUcZ7f8NsQVzVB6DPUAoVwwysDMjJ2XA426RA1tSYLy60sjbgNS+DXjQYMLVh+LHP7u6PCMrQ8hfBEV7nxx09hI9DhM0NXnmnGt8CKoiWahkQcsvjRLFO5odvOnpqCWUfINHZ8AUC2D7bD9s7nSv9eD6zZh+T2yYwIDAQAB;"
crm	60	IN	A	91.215.216.58
domain.bg	60	IN	MX	0	mail.domain.bg.
mail	60	IN	A	91.215.216.58

Also I check DNS zone for this domain in WHM and there everything looks good again (see the attachment). After all check I make a 3rd check via terminal on my computer and few other computers but there is a problem with TTL. When I run "dig domain.bg" the result is not what I expect.

; <<>> DiG 9.3.6-P1-RedHat-9.3.6-20.P1.el5_8.6 <<>> domain.bg
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 21100
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 2

;; QUESTION SECTION:
;domain.bg.		IN	A

;; ANSWER SECTION:
balkanbeauty.bg.	14400	IN	A	193.107.70.82

;; AUTHORITY SECTION:
domain.bg.	86400	IN	NS	ns2-70-82.superdnsserver.net.
domain.bg.	86400	IN	NS	ns1-70-82.superdnsserver.net.

;; ADDITIONAL SECTION:
ns1-70-82.superdnsserver.net. 636 IN	A	193.107.70.82
ns2-70-82.superdnsserver.net. 636 IN	A	193.107.70.82

;; Query time: 0 msec
;; SERVER: 195.191.148.254#53(195.191.148.254)
;; WHEN: Thu Mar 13 10:48:50 2014
;; MSG SIZE  rcvd: 147

I think my network may had a cache and check from other of our servers and the result is the same. After that I make a check from /http://www.dnsqueries.com/en/domain_check.php here and the result is the same.

So where is the problem?

I look at these topic before intervention:
http://forums.cpanel.net/f5/guide-transferring-all-accounts-new-server-18294.html
http://forums.cpanel.net/f5/how-incremental-serial-number-dns-zone-211732.html
Also few other topics from this forum and other forums.

Best regards,
Mirchev Ideas Team

 

[mirchevideas]

Hi,

I solved this problem and you can close this topic. What I do:
1. Log on the server as root via SSH
2. cp -a /var/named/chroot/var/named/ /var/name/chroot/var/named-backup
3. cd /var/named/chroot/var/named/
4. perl -pi -e "s/14400/60/g" *.db
5. perl -pi -e "s/86400/60/g" *.db
6. sed -i s/'[0-9]\{10\}'/`date +%Y%m%d%H`/ *.db
7. rndc reload
8. service named restart

And now everything looks great

Best regards,
Mirchev Ideas Team

 

[cPanelMichael]

Hello

I am happy to see that you were able to resolve the issue. Thank you for updating us with the outcome.

 

[Havri]

Hello,

Sorry for posting on this resolved topic. Just wanted to post an actual bash script that uses mirchevideas's method to change TTLs and serials on the local server + syncing with the DNS cluster using main_domain file from /var/cpanel/userdata/:

#!/bin/bash

now=$(date +"%m_%d_%Y_%H_%M")

    echo "Making backup of DNS zones"

    cp -R /var/named /var/named_bak_$now

  sleep 2;

    echo "Changing TTL values"
## These are our custom values. Everybody should change these values to fit their needs. If you have 86400, you should set the below settings accordingly.
## From 1800 -> 300; From 3600 -> 240

        perl -pi -e "s/1800/300/g" /var/named/*.db
        perl -pi -e "s/3600/240/g" /var/named/*.db

  sleep 3;

    echo "Changing zone serial date"

        sed -i s/'[0-9]\{10\}'/`date +%Y%m%d%H`/ /var/named/*.db

  sleep 3;

        rndc reload

        service named restart

  sleep 3;

    echo "Syncing modified zones to DNS Cluster"
       
        grep -r 'main_domain:' /var/cpanel/userdata/ | awk '{print $2}' | sort |
                while read DOMLIST;
                   do
                     echo "Now syncing ${DOMLIST} zones..";
                        /scripts/dnscluster synczone ${DOMLIST};
  sleep 2;   
                   done

echo "All done! Just to be sure, please check values in the DNS Sync-only servers."

Thank you.