Providing Redundant DNS Servers
Many registrars require the registrant to provide the addresses of at least two separate name servers. The purpose of this is to achieve some degree of redundancy - when a domain name is served by two name servers, name resolution will continue even if one of them goes down for some reason. A single-server installation of Plesk is unable to satisfy this requirement for reason that it is running a single DNS server instance. However, if you want to provide redundant DNS servers to your customers, there are a number of options available to you:
Configure at least two IP addresses on your Plesk server
This will enable you to add a second NS record to the server-wide DNS template. To do so, follow these steps:
- Go to the Tools and Settings > DNS Template.
- Add the following entries to the DNS template:
- ns1.<domain> A *the first IP address*
- ns2.<domain> A *the second IP address*
- <domain> NS ns1.<domain>
- <domain> NS ns2.<domain>
Note that while this setup satisfies the requirements of most registrars, it does not in fact provide redundancy. Should the Plesk name server stop responding, name resolution for the domains hosted on the server will stop.
Use cloud services
For example, you can run DNS in Amazon Cloud with help of the “Amazon Route 53” extension. Setting up the extension is described in detail in the Integration with Amazon Route 53 subsection, found in the Using External DNS Servers section.
Use an extra server
Another option is to use an extra server as a slave name server with the help of the “Slave DNS manager” extension for Plesk. You can download the extension here: plesk.com/extensions/slave-dns-manager.
Note that if you are running two or more Plesk servers, you can use them as slave name servers for each other. For example, if you have two Plesk servers named host1.com and host2.com, you can use host1.com as the master name server for domains hosted on that server, and host2.com as a slave name server (and vice versa).