Search the Community
Showing results for tags 'corruption'.
This is not a duplicate topic. I created a post here: http://forums.retrospect.com/index.php?/topic/150552-retrospect-configuration-gets-corrupted-too-easily/ with a solution for other users. This post is just for the bug report. The Config80.dat file is too fragile and does not survive forced power downs on the backup server. The current method of saving the configuration should be replaced with something that is more robust. The configuration file is also large - it's 80MB in my case - which suggests that maybe its design should be refactored or else the content split up into separate files so that the basic server configuration is less likely to get corrupted.
I have roughly 25 scripts that I use to automagically do the backups, and rotate the backups at the beginning/end of each month. I use proactive, regular "backup" scripts, and dataset copy scripts. Today, I was checking on things, and noted that 3 of the scripts were marked with a red X. One had lost its source media set, and two had lost their destination sets (and their schedules, too) I'm quite sure that I did not make these changes. The proactive backup script that was damaged is supposed to run a backup every day. It had not un its backup since the 15th of November. This configuration has been running happily for many weeks. I fixed it, but it is disturbing that the config can silently and spontaneously change.