Auto repair of corrupted backup sets

[awnews]

For the last few days I've been getting round-the-clock repeated emails reports showing the following error:

 

>From Retrospect: Script "JL_RMK5_WS1" failed during automatic execution, error -1127 (data error detected in file using cyclic redundancy check). Please launch Retrospect and check the log for details.

 

I finally had to make a trek across town to the Retro server and manually recycle the backup to clear this "cyclic redundancy check" caused by ??. I don't understand why I had to do this--surely the "automatically repair corrupted backup sets" setting should have taken care of this and fixed things without manual intervention. Ohhh, that's right, there *ISN'T* any "automatically repair corrupted backup sets" feature or setting....

 

+ Retrospect version 7.0.326 (Multi-Server)

+ Retrospect Update, version 7.0.12.105

[mlts22]

I'd like to see a chkdsk or fsck like function as well, especially for if the backup server has an issue, and crashes during a backup, leaving the backup sets it was accessing at the time in an uncertain state.

 

One idea, but it would slow things down, would to have Retrospect use a transaction-based system, journalling all writes atomically, so if a crash did happen, the backup set would be at some known good state.