j5 and its component parts should be set up on both the production and standby environment to be as similar as possible. This will make failover to the standby system as simple and predictable as possible.
File system replication must be used to ensure that data on disk (e.g. attachments uploaded by users) is kept in sync between the production server and the standby server. File system replication can also be used to ensure that the application files and configuration is kept in sync across the servers.
There are many tools and methods available for backing up files and directories. This document refers to the program robocopy, which is installed as part of Windows, and is designed to keep directories and files synchronized across different servers. Sample robocopy scripts are provided in the Example robocopy scripts section of this document.
More information about robocopy can be found at:
To replicate the j5 file system, the following steps are required:
Identical versions of the j5 system must be installed on production and standby environments, prior to setting up file system replication. (It is not sufficient to install on the production server and rely on file system replication to mirror the installation on the standby server, since that will not create the services, for instance.)
If the process of installing the j5 system on the standby server automatically creates the j5 services, these should be set for Manual startup instead of Automatic startup.
The backup user must have write permissions to the folders on both standby and production servers. It may be best to run the robocopy job as the j5 service user.
A scheduled job must copy the j5 system files onto the corresponding folder on the standby server.
A script must be available to copy new files back from the standby server to the production server. This is used in the failback process (once the production server is ready to be brought back online).