Question on Scheduled Incremental Backups

Hey guys, I have a bit of a lengthy question, well, short question but lengthy explanation.

When you do a standard back with a method of incremental, firstly, you do a full backup as you know (lets call the file FULLBACK). However, when you come back to do the next back up this time you actually choose incremental as the method. Upon completion of this backup another backup file is created (now lets call this one FULLBACK1) and then when you want to do the next incremental backup you would choose FULLBACK1 as the target. Now when that completes it would produce another file in the backup chain, FULLBACK2. Now lets say next week you wish to do another backup, this time your target file would be FULLBACK2, so the process would then produce FULLBACK3, and so on..

However my question is.. if you start fresh (assuming the full backup has already been made) you would schedule an incremental backup using the target location as FULLBACK1, and from then on out for whatever date you set it to backup on it would know to use only the target FULLBACK1 on this day. How would the scheduled task know to point to the newly created back up archive each time it is ran? Because for an incremental backup to work you have to choose the latest created archive, in which case.. since this is a scheduled task, it wouldn't know this and FULLBACK1 would remain as the target.

Any incite welcome.

 

"Because for an incremental backup to work you have to choose the latest created archive"

This is not correct - you can (and I do) choose the FULL backup image when creating a scheduled INC task.

 

Really? The little "quick hint" that popped up during my first INC backup said to choose the last archive in the chain as the target.

 

If you backed up on 2-9-08 and did incs every day 2-10 through 2-15, then on 2-16 if you did an inc based on the 2-15 inc, it would/should capture the changes since 2-15. If you based it on the 2-9 file it would have all the changes since 2-9. Either way you'd have all the changes since 2-9, just with a bigger inc on 2-16. So a recommended practice would be to base each inc on the last previous inc. If you're doing diffs, it doesn't matter, each diff will automatically have all changes since the full backup.

 

Sorry, but I think your logic is flawed.
You cannot look at an INC without also looking at the FULL on which it was based.
Basing an INC on the FULL will look at the FULL and all previous INCs (otherwise it would be a DIFF). Basing an INC on an existing INC will take into account that INC, the FULL and all other INCs as well. TI sees all these files as a single set.

“So a recommended practice would be to base each inc on the last previous inc”
So, how does one schedule a task based on an image that has yet to be created?

 

Which is my question

 

One does not, because it is not required. Set the task up based on the name of the FULL image.

 

HOw do you get form one to the other? Whyt would the last previously created file be "yet to be created"?

 

I misspoke about any diff between using a full or any associated inc as a base. Provided the full and incs are in the same directory, ATI will know that they are all parts of a set -- so you can point at any of them, and the the result is the same.

Per the user manual:
If you are going to append an incremental or differential file to an existing archive, select
any of the target archive files. It doesn't matter which one you select if the files are stored in
the same folder, as the program recognizes them as a single archive. If you stored the files
on several removable disks, use the latest archive file; otherwise, restoration problems might
occur.

 

My point was this –
If you run a FULL on Sunday and then create a scheduled task to run an INC on Mon, Wed & Fri, how could you possibly tell the Fri run to use Wed’s image since it has yet to be created?
You don’t; you tell it to use the FULL which HAS been created.

 

Yes however that contradicts what the actual program says in it's tool tips.. regaurdless I believe this is my answer. Thanks.

 

Not really a contradiction.
You can choose the FULL or any previous INC. The “tip” chose a previous INC.

 

No I see you're problem.

I'm pretty sure, when you create a task fopr a inc backup, you can specify a name to use and a location/directory to save in and it will make an inc if a full exists or a full if one doesn't. Anyhow that's what happens when I created a task for a scheduled an inc and ran the task a few times a few minutes ago.

So you could create an inc schedule without having had already made any inc or full first.