How we developed our DNS manager
Today we will talk about our latest innovation - the DNS-manager. It is based on the ANS solution and is already available to all our customers.
/ Photo by Keoni Cabral CC
We have approached the implementation of this functionality as we have further developed and improved other service systems of our IaaS provider. Previously, our clients could manage DNS only with the help of a support service - it was necessary to contact our specialists, who made all the necessary changes. Despite the operational work of our team in fulfilling such requests, automation did not take long to wait.
')
One of our main principles of work is the provision of tools that even those who have not previously encountered the IT component can use without problems (by the way, we are developing thematic guides on the add-on of various systems for additional user support). We have tried to follow this ideology in this case.
At the design stage, we analyzed various DNS server implementations. Among them were BIND, PowerDNS, Knot and NSD. All of them required serious improvement, bringing to a state understandable for the user and solving a number of security problems.
As a result, we opted for the ANS Carrier-Grade DNS Appliance product. It can be called a profile solution for IaaS and PaaS providers. In terms of performance, features for configuration and logging for all arranged. In addition to the main features, we also used the API of this solution.
In working condition, the implementation of the DNS manager allows you to make the necessary changes to the 1cloud control panel. Saving the changes triggers a new task, which is processed by the service and transferred to the DNS management server using the API.
From the user's point of view, everything is very simple. The example in the screenshot below.
Then you can create any new entry for the zone by filling out a simple form:
Automating the management of DNS records is only the first step in a series of planned work to continuously improve our services. At the moment we are considering automating the recovery process from the backup, the system of payment only for the resources used and autoscaling with the critical load on the already reserved capacity.
/ Photo by Keoni Cabral CCWe have approached the implementation of this functionality as we have further developed and improved other service systems of our IaaS provider. Previously, our clients could manage DNS only with the help of a support service - it was necessary to contact our specialists, who made all the necessary changes. Despite the operational work of our team in fulfilling such requests, automation did not take long to wait.
')
One of our main principles of work is the provision of tools that even those who have not previously encountered the IT component can use without problems (by the way, we are developing thematic guides on the add-on of various systems for additional user support). We have tried to follow this ideology in this case.
At the design stage, we analyzed various DNS server implementations. Among them were BIND, PowerDNS, Knot and NSD. All of them required serious improvement, bringing to a state understandable for the user and solving a number of security problems.
As a result, we opted for the ANS Carrier-Grade DNS Appliance product. It can be called a profile solution for IaaS and PaaS providers. In terms of performance, features for configuration and logging for all arranged. In addition to the main features, we also used the API of this solution.
In working condition, the implementation of the DNS manager allows you to make the necessary changes to the 1cloud control panel. Saving the changes triggers a new task, which is processed by the service and transferred to the DNS management server using the API.
From the user's point of view, everything is very simple. The example in the screenshot below.
Then you can create any new entry for the zone by filling out a simple form:
Automating the management of DNS records is only the first step in a series of planned work to continuously improve our services. At the moment we are considering automating the recovery process from the backup, the system of payment only for the resources used and autoscaling with the critical load on the already reserved capacity.
Source: https://habr.com/ru/post/261139/
All Articles