How to backup the server
It is essential that the database be backed up on a regular basis in accordance with your company’s data protection policies.
In order to backup a server you must
1) stop the server service
2) backup the contents of dataroot
3) restart the server service
Notes
- The location of dataroot is described in this knowledgebase article.
- A typical dataroot folder structure looks like the following.
Note that the folder …/db/bkup/ contains backups of the database taken during normal use. This folder can safely be excluded from a regular backup cycle to save time and storage space. These files allow for a quick recovery in the case of a database file corruption. This is not something that has ever occurred in practice.
3. The following commands can be used to stop and start the server service from a script. Administrator privileges are required.
sc stop cis-1
sc start cis-1
4. If restoring a database to different server you should consider excluding the following files from this restore process.
cis.ini
licence.txt
Both these files contain information potentially related to the system from which it originates copying it to another system might cause a malfunction.
In the event either of these files are damaged they can be reinstated by running the AMF product manger (install a product) and entering the license key for the server.
5. In order to minimise downtime we suggest that a local temporary copy be made whilst the server is stopped in cases were the copy to backup media has a significant latency.
6. It is theoretically possible to take a copy of the database without stopping the server, by utilising the most recent hourly backup file from the bkup folder. However as the external files (BOMs AMLs and other managed files) change independently the database the restored data will not reflect a specific point in time. In many cases this will be sufficient, but in practice the downtime related the copy period is likely to be short enough to make this approach unnecessary.