What happens when 'Do not schedule' is chosen?

Hans, welcome to these public User Forums.

If you set any task to 'Do not schedule' then the only backups made will be when you have started them manually by clicking on the Backup now button. 

The number of versions created etc, is controlled by the settings you make on the task Backup Scheme page in the Options.

Steve Smith wrote:

Hans, welcome to these public User Forums.

If you set any task to 'Do not schedule' then the only backups made will be when you have started them manually by clicking on the Backup now button. 

The number of versions created etc, is controlled by the settings you make on the task Backup Scheme page in the Options.

Thanks Steve for your answer.

This is a little disturbing for me because it is not what I expected. Let me explain what I want my backup program to do. I have a 1 TB internal data harddisk and a 2 TB external backup harddisk. I want 1 full backup and the remaining space to be filled up with incremental backups until al the space is used up. Then I expect ATI to delete the oldest backups as much as needed to complete the next backup job. My question is: how can I set this up?

Thanks in advance

Hans 

I have a 1 TB internal data harddisk and a 2 TB external backup harddisk. I want 1 full backup and the remaining space to be filled up with incremental backups until al the space is used up. Then I expect ATI to delete the oldest backups as much as needed to complete the next backup job. My question is: how can I set this up?

Hans, sorry but ATI does not work in the way that you are wanting to use it.

Automatic cleanup operates on complete version chains, i.e. a Full backup plus all associated incremental backup files that share the same _b?_ number in the file name.

Nothing will be deleted until a new full backup file for the next version chain _b?+1_ has been successfully created.  This means that you always need to have sufficient free space on your destination backup drive to hold that new full backup file.  So you cannot make 1 full backup and then fill the remaining space with incremental backups.  The oldest backup in such a scenario would be the initial Full backup file which cannot be deleted without also deleting all the associated incremental files too, as those files are useless without the initial full they are based on.

Thanks again for your quick explanation. You know, my worst nightmare is when I am working on corrupted data without knowing it. I want a possibility to trace back to the uncorrupted data, say 2 or 3 Months back in time (just a guess). Do you think this is possible?

Hans

Hans, a lot here depends on what your data is that you are wanting to protect from corruption.

I would suggest dividing your data into more manageable portions, i.e. make separate backups of that data, one a backup of your main OS partition & applications, another backup of your most important data (files & folders).  The advantage of this approach is that these backups can have different schedules based on how often change occurs, so your OS backup would only be needed when significant change has occurred such as a Windows Update or new application install etc, whereas your data backup can be much more frequent to cope with data changes.

You backup scheme should include multiple backup version chains so that automatic cleanup can be used to manage the files held on your backup drive.

Okay. The part of separate backup scheme's is clear to me. Just to get "definitions" right: a backup is a full backup plus the following incremental backups; a backupversion is the next full backup plus the following incremental backups. Is this right? Why is there a default upper limit of 20 versions?

I started the following post before reading Steve's latest post.  It doesn't say much he didn't but I'll post it anyway.

Hans Maas wrote:

... my worst nightmare is when I am working on corrupted data without knowing it. I want a possibility to trace back to the uncorrupted data, say 2 or 3 Months back in time (just a guess). Do you think this is possible?

This is a problem a lot of us face.  There are (at least) two reasons to take backups:

1. to be able to recover data and/or the operating system from a catastrophic error
2. to create a historical archive in order to access old copies of important data

I think it is best to have separate backup schemes to address these two needs.  For the 1st you really need nothing other than the most current backup.  The 2nd obviously requires multiple backups over time.  (Full system backups with multiple generations obviously combine the two schemes, but saving many generations of full system backups takes a lot of space and results in multiple copies of a lot of unchanging data.)

I would recommend identifying the data that you really need archived and create a "files and folders" backup scheme just for that data.  Save as many generations as your space allows.  But keep in mind you must have room at your destination or your backup will fail, and backup space is reclaimed only at the completion of a successful backup.  It might be best to have different destination drives for archived data and recovery data.

Just to get "definitions" right: a backup is a full backup plus the following incremental backups; a backupversion is the next full backup plus the following incremental backups. Is this right? Why is there a default upper limit of 20 versions?

Hans, all backups require an initial Full backup file to be created, after which you can create either Incremental or Differential type backup files.  A backup version chain comprises 1 x Full plus as many Incremental or Differential files you create prior to making the next Full backup.

With Incremental backup chains, all the files are required / needed for any recovery action.

With Differential backup chains, only the initial Full plus one of the Differential files are needed.

Incremental = the changes to source data since the prior backup (full or incremental)
Differential = the changes to source data since the initial full backup.

There is no default upper limit of 20 versions other than the default settings for Cloud backup tasks. The actual limit is determined by the amount of storage you have on your backup location.

Hans Maas wrote:

Thanks again for your quick explanation. You know, my worst nightmare is when I am working on corrupted data without knowing it. I want a possibility to trace back to the uncorrupted data, say 2 or 3 Months back in time (just a guess). Do you think this is possible?

Hans

I fully share your fear. I've dealt with that very issue. Actually, I've dealt with customers who were suicidal because they were dealing with that very issue.

Here's how I handle this. I have one backup drive on which I store monthly full backups. I have another drive on which I store incremental chains of thirty-two daily versions. The drive for the full backups can store a dozen full backups, which gets me twelve months worth. The drive for the incremental backups could store about four chains, but I have it set to keep only three. 

In this way, I can go back to any day for the last three months; and I can go back to any month of the last year.

I settled on using two different backup drives for a couple of reasons: If either drive fails, I won't lose all of my backups. And while the size of the monthly backups is fairly predictable, the incrementals can suddenly get big if you refresh a large number of files. So I know pretty accurately how many generations of monthly backups I can keep, and take my chances with the incrementals.

Now, this is something I can "handle". I was just wandering: do you use this 2x1TB raid 1 for your backup system?

No, the RAID 1 setup is my working drive. I like the idea that if one drive fails, the other just keeps plugging along until I can get a replacement.

Not long ago I had an external drive fail. It took my incremental backups with it, but my monthly backups were safe.

I also take one other precaution: I back up my important data files to the cloud every night. If something goes horribly wrong (a fire, for example), I might see a big pile of burnt and twisted metal where my computer used to be, and two smaller piles where my external drives used to be. It's nice to know that my tax records are someplace offsite.

In case you hadn't guessed, I used to do single point of failure analysis for some pretty big outfits.

Okay, thanks for your answer Jerry. Then what is the size of your backupdisk(s) if you can restore up to 1 months back? And what is the size of the disk for your incrementals?

Thanks in advance

Hans

My full backups are between 250 and 300GB. A full incremental chain can be around a terabyte, although it is usually much smaller. That depends on how much stuff has changed from one day to the next.

Hello

How can I start a new full backup? When I do this will my old full back-up including its incrementals be deleted/cleaned?

Hans

Hans Maas wrote:

Hello

How can I start a new full backup? When I do this will my old full back-up including its incrementals be deleted/cleaned?

Hans

How can I start a new full backup? When I do this will my old full back-up including its incrementals be deleted/cleaned?

Hans, the answer here really depends on the settings you have for your Backup Scheme and automatic cleanup options, along with how much space you have on your backup destination and the size of your backup chain.

Steve

I think I didn't express my question very well, because I don't understand this to be an answer to my question.

I am trying to figure out how ATI works. I can see in the Activity tab that a full backup has been made and a series of incrementals after that. Unfortunately the cleanup begins after a maximum of 20 versions  which -imo- is an arbitrary number and makes no sense when (a) file(s) are/is changed on a regular basis. Writing an article for a magazine or a book or a thesis, for instance, might take more than 20 versions and when this article needs to be peer-reviewed then a span of 5-6 months can happen as well. So,- in theory- backupspace is not a factor in this question because there are so many cloud-solutions. 

With that in mind can you answer my question?

Thanks in advance

Hans

I am trying to figure out how ATI works. I can see in the Activity tab that a full backup has been made and a series of incrementals after that. Unfortunately the cleanup begins after a maximum of 20 versions  which -imo- is an arbitrary number and makes no sense when (a) file(s) are/is changed on a regular basis. Writing an article for a magazine or a book or a thesis, for instance, might take more than 20 versions and when this article needs to be peer-reviewed then a span of 5-6 months can happen as well. So,- in theory- backupspace is not a factor in this question because there are so many cloud-solutions. 

With that in mind can you answer my question?

Hans, thank you for clarifying your question.

There are different configuration factors involved here that need to be understood.

For a Backup Scheme, the normal default would be to create a new Full Backup after 5 Incremental backups have been created.  This would create a Backup Version Chain comprising of 1 x Full backup and 5 x Incremental backup files.

Automatic Cleanup works on completed Backup Version Chains and will not remove any old chains until a new Backup Version Chain has been started by the creation of a new Full Backup file.

Along with the above definitions, you also have the different types of Backups that can be created, i.e. Entire PC (capturing all data from all connected drives); Disks & Partitions (capturing all data from one or more selected disks & associated partitions); and Files & Folders (capturing all or selected data from one or multiple folders on one or more disks / partitions).

Applying the above to your example of writing an article for a magazine or book or thesis, then the obvious Backup type to choose would be of Files & Folders, where the folders / files selected are those that hold your documents related to this task. 

Having chosen to create a Files & Folders backup, you can then further choose the Backup Scheme to use for that data capture, so this could be a good candidate for choosing Non Stop Backup scheduled task, where new data can be captured at a minimum of 5 minute intervals.

If you don't need such an intensive backup scheme, then choosing an Incremental backup scheme with a suitable number of Incremental backups being created before then creating a new Full backup should be considered, where the size of data involved can help influence the values being chosen.  This is then used in conjunction with the Schedule options for how often the task will be run, so you could run it every hour, (or 2, 3, 4 hours etc) or twice a day, once a day etc.

Having established your Backup type (Files & Folders) and Backup Scheme (Incremental) and Schedule - the next choice is for Automatic Cleanup.  This again is influenced by the size of data and the retention period needed for the data.

Example time:

If your Files & Folders backup creates 23 Incremental files then starts a new Full Backup Version Chain, and is run once an hour every day (24 x 7), then you would create one new Version Chain every day.

Automatic Cleanup could be configured to Store no more than 180 recent version chains, thus keeping 6 months of backup chains.

The size of data then determines what size drive is needed to store these backups!

Thanks a lot, Steve, for this ample explanation.

I need to study it, alongside the manual to figure out what to do next and eventually change my present backup scheme.

You helped me a lot, so thanks again.

Hans