DNS Errors
1. rndc: connect failed: 127.0.0.1#953: connection refused
This error may appear on Centos 6.3+ when named is unable to bind to port 953 for rndc. This is caused by the named init script not being configured properly for "portreserve" and can be fixed using the following steps:
# mv -v /etc/init.d/named /etc/init.d/named.prfixbak
# yum -y reinstall bind
# service named stop
# service portreserve restart
# service named start
2. Using named-checkzone to check for errors
named-checkzone will check a given DNS zone for errors.
# named-checkzone avinash.com /var/named/avinash.com.db
3. Flush local DNS cache
The local DNS cache can be flushed, thereby forcing an actual domain name lookup. You can do it using following steps:
In Linux: Open up a root terminal window (ctrl T in gnome). Type the following command and hit enter.
# /etc/init.d/nscd restart
Restart your application (e.g. browser or email).
In Windows: Run the following on command prompt:
$ ipconfig/flushdns
Restart your application (e.g. browser or email).
4. rndc errors
If rndc is reporting the following error:
# /scripts/fixndc
5. Custom Nameservers never propagate
The custom nameservers (ns1/ns2.domain.tld) and DNS data are not propagating after setting up custom nameservers.
This is due to having “allow-transfer” declared twice in the named.conf. You will want to leave allow-transfer enabled for the loopback device (127.0.0.1), so edit the file and find the allow- transfer line for “none” and comment it out, then save the file.
Now, restart named with service named restart. This requires a full restart of named, a ‘reload’ will not be sufficient. Remember, whenever you make a change to the configuration file (named.conf), you have to restart named.
# mv -v /etc/init.d/named /etc/init.d/named.prfixbak
# yum -y reinstall bind
# service named stop
# service portreserve restart
# service named start
2. Using named-checkzone to check for errors
# named-checkzone avinash.com /var/named/avinash.com.db
3. Flush local DNS cache
The local DNS cache can be flushed, thereby forcing an actual domain name lookup. You can do it using following steps:
In Linux: Open up a root terminal window (ctrl T in gnome). Type the following command and hit enter.
# /etc/init.d/nscd restart
Restart your application (e.g. browser or email).
In Windows: Run the following on command prompt:
$ ipconfig/flushdns
Restart your application (e.g. browser or email).
4. rndc errors
If rndc is reporting the following error:
rndc: connection to remote host closed
This may indicate that the remote server is using an older version of
the command protocol, this host is not authorized to connect,
or the key is invalid.
This is caused by the incorrect key specified in the "/etc/rndc.key". To correct this you can generate a new one by running the following script:# /scripts/fixndc
5. Custom Nameservers never propagate
This is due to having “allow-transfer” declared twice in the named.conf. You will want to leave allow-transfer enabled for the loopback device (127.0.0.1), so edit the file and find the allow- transfer line for “none” and comment it out, then save the file.
Now, restart named with service named restart. This requires a full restart of named, a ‘reload’ will not be sufficient. Remember, whenever you make a change to the configuration file (named.conf), you have to restart named.
Comments
Post a Comment