Problem with log backups
I'm using Red Gate SQL Backup to perform many tasks, among them are log backups for some databases. Today I found something rather strange.
Log backup failed because there was not enough space on the destination drive - OK :-), such situtaion should never happend (that is - to some extent - my fault), but the problem is not in the fact that the backup failed - but in that the next log backup performed when there was a lot of space, which completed successfuly, contains LSNs newer than the last LSN in the previous successful log backup - so there is a gap, and as a result log backup sequence is broken.
According to MS and in case of the normal SQL command BACKUP LOG...:
The log is backed up from the last successfully executed log backup to the current end of the log.
Can anyone explain that behaviour?
I'm more than happy with Red Gate SQL Backup, and the process that failed today worked without an issue for over a year (in this case as a part of the custom log shipping scenario - that's why the gap in the sequence is so important), but I want to be sure, what I may or may not expect in the case when the log backup will fail in the future for any reason.
I'm posting this message to the forum about Red Gate SQL Backup 5 (because in this case it was version 5.x) , but maybe that behaviour changed in the version 6.x?
Best regards and thanks for any suggestion/responses