Good day to all. I would like to congradulate you all on a brilliant product. I have been watching this list for some time as I find the topics most informative. I have a carefully crafted question that I would like to ask you guys. But first some info to help give you a better picture. I am using a RH 9.0 box and the latest rsync to backup different servers to an offsite data facility. The server being backed up are also RH 9 with the latest rsync. I have acquired so enough customers that space on the rsync server is running low and I am looking for ways to curb the space needed to update the customers "account". I have rsync run from the customer's server to the rsync box nightly after the end of day process has run and activity has stopped. after the system is synced up with the customer, the system makes an archive of that customer's data for that day from the updated data. The problem is, that the "pool" of data that is used to create the archive is in some cases, huge. Is there a way to make rsync sync up the changes without actually having the data on the server ? Thank you very much for any help that you can render from this message. If there is anything else I could tell you to clear up any uncertainty, please let me know. Cheers; C -- Clint Boggio <clint.boggio@hoganhardwoods.com> Hogan Hardwoods & Moulding
You can rsync back over the same data, and only the changes will be brought over. If you're looking for an incremental backup sort of a deal, use the --backup-dir option, which will save changed files over to a new directory, so you can reach back in history. The ability to roll the filesystem back to a specific data, including not having things there that were added later, is a bit stickier, but there are professional packages to handle that. ADSM(tradename has been changed to "TSM" now) is particularly sweet in that regard, and I hear the Veritas NetBackup is nice, too. Tim Conway Unix System Administration Contractor - IBM Global Services conway@us.ibm.com