>ACK that this is more sledge hammer than elegant. But it is a KISS approach where we are not dependant on all previous versions to be error free, so the worry of failing backup is minimized and the verification attempts as well ;-)
>
>I'd hate to be one of the people realizing that some backup step misfired after the need for backup data is pressing.
>One of my early brushes with such scenarios was in a now famous Italian bank here in F right after XMas and shortly before years end runs: during some building effort scheduled in the off-time a lot of the servers were physically impaired as well as backup storage "vault". The off-site redundant backup tapes had been freshly formated to be used again with that ending years data ;-)
Offsite synchronization is done manually since a while in the wait of a more fancy implementation. Whatever we do can only be better and this is also towards secondary sets of servers, into which, their importance level is somewhat considered lower than the production one. As far as the data, this is a full copy of the database on a daily basis so that is as is. The files on disk are images, banners and things like that.