Full PC backup from Windows or not?

Is the image you create from TIA launched from Windows and TIA launched from bootable CD different?

In my opinion yes: the backup produced in the first case it's not good, because Windows is active, and so the status you store is Windows-on, while it should be better to have Windows-off. Am I wrong?

So.. why does Acronis permit full backup of your computer from Windows?

Mhaxx

 

having make literally thousands of images from within windows I'm inclined to believe that there is no practical difference between a windows image an an external image.

 

Yeh theres no difference from my experience. These days all my images are made from bootable media as don't need to have ATI installed.

 

One of the useful features of TI is that it CAN create full system backups from within Windows.

 

You're entitled to your opinion, even if it's wrong. The fact is that backups produced from the Windows version of ATI do work, and can be restored. Many of us use them.

While you may not understand how it's technically possible, Acronis has obviously made it work.

It works.

 

Let's suppose there's an application which runs when you're performing a full backup of your computer from Windows and that it changes its status from "off" to "on" keeping the status information in a file and depending on it ("on" or "off") that application starts in different way when you launch it.

This means that backup stores the "on" status even if that image will be restored and launched from "off" mode, that is when you'll start that application after the restore its behaviour will be depending on "on" instead of "off" status and this could change what it does.

I think the only way Acronis performs the backup is to store all the files and configurations like they are in THAT moment, but (you know) the content of some files is different if Windows is running or not. Then I think you could have some (hidden?) problem if you launch Windows restored from an image produced with Windows "on".

It's only my opinion..


Thanks for your feedback,

Mhaxx

 

Yes, your opinion – but wrong.
There is a detailed explanation of how it works on the TI web site.
It takes a snapshot of the in use sectors and then steps in between the OS and file system. As it backs up the in use sectors it monitors what changes are taking place and makes the appropriate steps to include those changes.

 

Here: http://www.acronis.com/company/inpress/2003/01-ctr-disk-imaging.html

 

An ATI image made from within Windows is a "Point in time snapshot". Here is a description from Acronis about their snapshot technology.

Let's call this a "live" image. If you restore a live image then your machine is restored to the point in time when the image was created. Any changes made after the imaging process started will be lost, of course.

The only anomaly that I've run into with this method occurs when making images of a running Vista system. Since Vista keeps track of the system state as "running" or "shut down", when you restore a live image and first boot into it the Vista boot manager will detect the stored state as "running" and you will see a "Windows was not properly shut down" error message. If instead you create a Vista image from the recovery CD when the OS is shut down, then the system state will be detected as "shut down" on first boot after restoring the image and the error message will not be seen.

Other than this, Acronis "live" imaging works very well.

 

You may choose to believe your own opinion, even though many people with more experience than you, and Acronis documentation itself, contradict you. If you prefer to use only the bootable recovery CD to do your backups, that's fine. I don't know what your purpose is in continuing to post incorrect information, but feel free.

 

“Any changes made after the imaging process started will be lost, of course.”

No, I don’t believe that’s true. TI monitors all changes that take place while the backup is in progress and stores them in a buffer which later gets applied to the image.
That’s how I read it.

 

Interesting. I imaged from within Vista Home Premium and later restored it. When I booted, I didn't get any error message.

 

Weaz:

Was this the part that you read?

I believe what this means is that ATI lets the file system write to sectors on the disk that have not yet been included in the final image. But it also copies the sectors (before they get overwritten by new data) to a special buffer. It is this buffer that gets included in the image file.

I haven't run a test to confirm this, but I am pretty sure that any changes made after you start to create a live image in Windows will NOT be included in the image file.

 

ATI 10 or ATI 11? My ATI 10 does this.

 

I used ATI 11 on the Vista system.

 

Recovery CD Backups

I'm not doubting that Images created from within Windows are just as good as those created via the recovery CD. I've never used Windows to create an image, I always use the recovery CD. My reasoning: I believe the chances are better when creating an image with no open files on the drive.

 

Mark,

So, in other words, the backup is, in effect, finished almost the moment it begins. All that’s left to do is transfer the information into the image file.
Since no backup software could include changes made after the backup is complete, TI performs the way one would expect. The backup contains all the data as is was at the initiation/completion of the backup process.

 

Weaz:

I think it contains all of the information at the beginning of the image creation process, at the moment the "snapshot" is taken. An analogy is taking a huge, high-resolution photo with a digital camera. After you "snap" the picture, the contents of the picture are frozen in time. Afterwards, it may take a bit of time to process, compress, and transfer the digital image file to flash memory.

Again, I haven't tested to confirm this. One test would be to start the image creation process with TI in Windows, and while it's running, delete a file. When complete, mount or explore or restore the image. If the file is gone then my explanation is correct. If it's still there, then your explanation is correct.

 

This morning I ran a little test. Midway through the full backup of my system partition I opened Word and created a small test file and saved it to the Desktop. When the backup finished, I explored the image file and the test Word doc was NOT there.
So it seems that when you click Proceed (or the task is auto-launched via the scheduler) you are telling TI to “backup my data as it is right now, at this exact point in time”. TI does this while allowing changes to take place, but those changes will not be part of the backup file.
As I alluded to in an earlier post, the data that is to be included in a backup is determined at the very offset of the task. Changes are allowed to take place, but since the state of the data has already been established, the changes will not be included in the backup image. Again, no software would be expected to include changes made after the backup had finished. For all intents and purposes, a TI “backup” is finished almost as soon as it starts. The rest of the time is just stuffing the info into an image file.
It’s early, so if this makes no sense … yawn … sorry.

 

Your explanation makes sense; that's how I always assumed TI operated. Thanks for doing the test.

 

I work with mainframe apps that will buffer incoming changes if a backup/update is in progress and then later apply the changes. I was under the wrong assumption that TI worked that way. That is, during a backup, buffer incoming changes and then later apply them to the file system and the backup image.