The Importance of Monitoring Backups
Monitoring the success (or failure) of your backups is a critical part of maintaining a good restore point, should it be needed. Monitoring this process should involve all of the following on a regular basis:
- Checking backup logs for success and failure messages.
- Checking operating system logs for hardware and application software errors.
- Clearly labeling and rotating your backup media. Separate tapes per day, for month-end, for quarterly processing, and annual year-end is desirable. A set of about a dozen is a good idea. We recommend the you start with two boxes of ten tapes and develop you usage rotation.
Don't trust your backup plan if it involves reusing tapes too frequently. Remember that if your backup fails the tape in the tape then is not usable and the data on the disk is not backed up. Having a spare tape if one fails is a good hedge on recovering from failures. DO NOT just take another tape that is i the rotation and use it telling yourself you will get some tapes later. Using a tape out of sequence leads to cascading problems or losing track of what is on the tapes you have. It is far better to introduce a replacement tape that was NOT in service.
- Periodically checking integrity of your media
- Storing media in a fireproof safe.
- Taking backup media off-site for safe storage. Failure to do the above can create situations where a user has little or nothing to recover from.
As an example:
I had a recent issue with a customer reporting an abnormal number of corruption problems in temporary transaction files during posting operations. This was also general slowness on the PC. That behavior is typical of what we see when a hard drive is about to go permanently south.
However, in working with their local MIS personnel, the main issue was determined to be – the hard drive had less than 45 MB of disk space. Apparently, the backup software was configured incorrectly and was backing-up rotational file sets to the local hard drive, not the external drive. This filled the hard drive up, and caused the nightly operations to fail with to insufficient disk space. In this particular case, it meant that nothing had been backed-up for over 4 months.
To add insult to injury, after the disk space was freed-up, without making a current backup, the end-user decided to purge the Vendor History data. Getting distracted by a phone call, a 12/31/2010 cut-off was input, which purged ALL of the vendor history. The only way to get around this was manual re-entry of A/P data from the last good backup forward…
If you need assistance with checking, setting-up, and/or monitoring your backups, please contact the CCS Retail Systems Support Department.