Controlling a high availability computing system

Patent metadata

Field	Value
Publication number	US-10069788-B1
Application number	US-201615084188-A
Country	US
Kind code	B1
Filing date	Mar 29, 2016
Priority date	Mar 29, 2016
Publication date	Sep 4, 2018
Grant date	Sep 4, 2018

Abstract

A method for providing high availability computing service. The method comprises monitoring a high availability computing service by monitoring applications, each monitoring application executing on a different computer system, one of the monitoring applications executing in a primary role, and the remaining monitoring applications executing in a secondary role. The method further comprises writing a last touched time value into a DNS resource record of a domain name system (DNS) server by the monitoring application executing in the primary role, reading the last touched time value in the DNS resource record of the DNS server by each of the monitoring applications executing in the secondary role. The method comprises determining by one of the monitoring applications executing in the secondary role to assume the primary role based on determining that the last touched time value in the DNS resource record of the DNS server exceeds a predetermined age threshold.

First claim

What is claimed is:

1. A method of providing a high availability computing service, comprising:
monitoring a first computer system and a second computer system by a monitoring application executing on a third computer system, wherein the monitoring application executing on the third computer system is acting in a primary role, where the first computer system and the second computer system execute the same service application in a redundant configuration and wherein each redundant configuration is referenced by a common domain name;
monitoring the first computer system and the second computer system by a monitoring application executing on a fourth computer system, wherein the monitoring application executing on the fourth computer system is acting in a secondary role;
writing to a DNS resource record by the monitoring application having the primary role to a domain name system (DNS) server, where the DNS resource record is a DNS TXT record and indicates a last touched time, wherein writing to the DNS TXT record is performed according to dynamic DNS;
determining by the monitoring application having the primary role that the service application on the first computer system is not available;
in response to determining the service application on the first computer system is not available, writing one of a DNS A record or a DNS AAAA record that associates an IP address of the second computer system to the common domain name by the monitoring application having the primary role to a plurality of DNS servers, whereby the monitoring application having the primary role fails over the service application to the second computer system;
monitoring the DNS TXT resource record in the DNS server by the monitoring application executing in the secondary role, wherein the monitoring comprises reading the last touched time in the DNS TXT record and determining an age of the last touched time by comparing a current time;
detecting by the monitoring application executing in the secondary role that the last touched time in the DNS TXT resource record in the DNS server is more than a predefined period of time old based on the monitoring; and
in response to detecting that the DNS TXT resource record last touched time is more than the predefined period of time old, writing to the DNS TXT resource record by the monitoring application executing on the fourth computer system to indicate that the monitoring application executing on the fourth computer system has assumed the primary role.

2. The method of claim 1 , wherein the service application is a network operation center (NOC) application.

3. The method of claim 1 , wherein the service application is an e-commerce on-line retail sales application.

4. The method of claim 1 , wherein the service application is an electrical power distribution grid monitoring and management application.

5. The method of claim 1 , wherein the first computer system comprises a plurality of different hosts and the second computer system comprises a plurality of different hosts.

6. The method of claim 5 , wherein the locations of the first computer system and the second computer system are different and provide geographical redundancy.

7. The method of claim 1 , wherein the monitoring application having the primary role writes to the DNS TXT resource record periodically.

8. The method of claim 7 , wherein the monitoring application having the primary role writes to the DNS TXT resource record about every one minute.

9. A method of providing a high availability computing service, comprising:
monitoring a high availability computing service by a plurality of monitoring applications, each of the plurality of monitoring applications executing on a different computer system, one of the plurality of the monitoring applications executing in a primary role, and the remaining monitoring applications executing in a secondary role;
writing a last touched time value into a DNS resource record of a domain name system (DNS) server by the monitoring application executing in the primary role, wherein the DNS resource record is a DNS TXT record, and wherein writing to the DNS TXT record is performed according to dynamic DNS;
reading the last touched time value in the DNS TXT resource record of the DNS server by each of the monitoring applications executing in the secondary role;
determining by one of the monitoring applications executing in the secondary role an age of the last touched time value in the DNS TXT resource record of the DNS server;
determining by the one of the monitoring applications executing in the secondary role that the last touched time value in the DNS TXT resource record of the DNS server exceeds a predetermined age threshold based on determining the age of the last touched time value;
in response to determining the last touched time value exceeds the predetermined age threshold, writing an identity of the one of the monitoring applications that determined the last touched time value exceeds the predetermined age threshold into a primary monitoring application identity field in the DNS TXT resource record of the DNS server, whereby the monitoring application executing in the secondary role transitions to executing in the primary role on the event of failure of the monitoring application that had been executing in the primary role.

10. The method of claim 9 , wherein the monitoring application executing in the primary role writes the last touched time value into the DNS TXT resource record of the DNS server periodically.

11. The method of claim 10 , wherein the predetermined age threshold is at least two periods of the periodic interval of writing the last touched time value.

12. The method of claim 10 , wherein the monitoring application executing in the primary role writes the last touched time value into the DNS TXT resource record of the DNS server about every 1 minute.

13. The method of claim 10 , wherein the monitoring application executing in the primary role writes the last touched time value into the DNS TXT resource record of the DNS server about every 5 minutes.

14. A high availability computing service system, comprising:
a plurality of application servers, each application server comprising:
an application server processor,
an application server non-transitory memory, and
a service application stored in the application server non-transitory memory that, when executed by the application server processor, receives and processes client requests addressed to a common domain name that is mapped by a domain name system (DNS) to an application server from among the plurality of application servers; and
a plurality of service monitoring servers, each service monitoring server comprising: a monitoring server processor,
a monitoring server non-transitory memory, and a monitoring application stored in the non-transitory memory that, when executed by the monitoring server processor, that is configured to operate in a primary monitoring mode and to operate in a secondary monitoring mode,
when operating in the primary mode, the monitoring application monitors the plurality of application servers, writes a current time value periodically to a last touched time value into a DNS resource record of a DNS server, and when one of the plurality of application servers fails, writes one of a DNS A record or a DNS AAAA record to a DNS server that associates an IP address of an active one of the plurality of application servers to a common domain name whereby the client requests to the failed application server are directed to an active application server, wherein the DNS resource record is a DNS TXT record, and wherein writing to t

Assignees

• Sprint Communications Co Lp

Inventors

• Earl Bryson
• Szymanski Todd

Classifications

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L41/12

  Discovery or management of network topologies · CPC title

• H04L43/0805

  by checking availability · CPC title

• H04L41/0609

  based on severity or priority · CPC title

• H04L43/0817

  by checking functioning · CPC title