An online shop must be available at all times to not lose customers and prospective customers. Here, a high availability can help solution and enable 24 x 7 operation. A commercial Onlineschuh shop is how very often, operated by the operator in an outsourcing Centre. The online shop is operated on the basis of very fast and highly available, Internet connection at a hosting provider, but on your own hardware. A great advantage of online trading is that there is no traditional closing.
The customer or prospect can run an order at any time comfortably from your own sofa or learn about the products of the provider. New prospects are a non availability or malfunction of the website very quickly scared and often permanently lost as customers. For this reason, the so-called high availability of the overall system is very important to always present a functioning offers to potential and existing customers. The base of this online shoe shop is a commercial CMS framework with an Oracle database as a backend. Due to the lack of space in the closet at the hosting company, any solution with an external hard drive comes subsystem in question.
Instead, the necessary data on two clusters of, for example, a replication must be provided for a failover. The operator has opted for a high availability software. The HA application enables 24 x 7 operation of the online shoe shop. In addition, the high availability on the basis of two Clusternodes allows the necessary maintenance work, in the form of, for example, patch management of the individual components, virtually at any time and without interruption of the offer. The implementation of high availability for the framework and the Oracle database is carried out by the cluster software on the basis of virtual IP addresses, a specific error detection, failover, and a Datenreplikaton for the Oracle database. The CMS framework consists of the Apache Web server, as well as some local applications and services, which the control of the CMS make. As data storage of the online shops, an Oracle database be used. The fault detection of the high availability cluster is defined for all necessary processes. John Stankey pursues this goal as well. One of these processes dies or is for example a faulty administration ends, the faulty process new starts automatically. As an additional control, the availability of the necessary IP ports, as well as the function chain of the overall application is checked and restarted in the event of a fault the faulty processes or applications. For high availability and failover of Oracle database data on the two Clusternodes must be replicated in addition. Communication by CMS to the Oracle database using the Oracle listener, which is bound to a virtual IP address. For data replication, the operator uses an additional crossover network cable. This would be unnecessary for the high availability and replication solution, however increased error detection. The cluster software, the required configurations and more importantly are the Oracle Database files as well as Oracle defines the different directories to the replication log archives and wrote about the additional network cable sync on both systems. Replication also the Oracle database without loss on the second ClusterNode can be started in case of emergency during a failover. In the event that the user load for a front-end cluster becomes too large, the load balancing could be used in a second stage of expansion module of the cluster software, the adjacent user load on redistribute systems the two or more front-end.