Validation question

How do I validate a specific incremental backup archive. Every time I try to validate a specific one (say #3) on summary it lists the very last incremental backup (#5)

 

You can't validate an individual incremental. TI validates the entire archive regardless of which part of the archive you select for validation.

Two sides of the argument whether or not it should do this. One side is that you only wanted to validate one part of the archive and you had validated the rest of it earlier. Other side says the archive is useless or partially useless unless the whole thing works so it all should be validated.

 

THe reason is because of the way Acronis implemented TI.

Several other well known backup programs create separate files for each logical drive, so, for those programs, as I recall, it's possible to verify a drive at a time. Let's call this the "drive based model".

For TI, Acronis implented a real file system mode. In a real file sysyem, there is really no such thing as a drive, rather, there are mount points. The concept of a drive is an artificial thingee in a real file system.

So TI keeps track of things over the entire file system, in particualar, all the files for a particular "drive" may be spread out over several .tib files.

One drawback to TI's implementaton is that the verify phase takes a lot longer than when using the "drive" based file model.

 

Regardless of the concept, the validation process doesn't care much, if any, about what the data in the archive represents. It just reads the file, re-constructs and compares the checksums contained in the file. There is no reason that this process couldn't be restricted to the file(s) associated with a selected incremental stage.

 

Only Acronis can cite the details of their implementation.
I may be that verifying certain checksums, e.g., to certify sanctiy of a directory, or a file system structure, is not self-contained in a .tib file.

 

Slightly off thread I or rather ATIH spent 2 hours this AM backing up a partition and then in 10 minutes verification told me it had failed. WHY?

I have spent a lot of time recently backing up multiple partition HDs and then when I needed them they failed to verify as well. Is there no automatic verification taking place when the backup is created and if not why not?!

Really very *********** in the UK.

 

I cursed Acronis for some time as I kept getting unverifiable backups. Then I tried running MEMTEST and found I had dud memory 8-{

On reflection it is quite reasonable that the data verification process should pick up memory faults more effectively than most other applications. I wish I had realised when I first saw the errors that it was possible TI might be being helpful rather than the opposite.

Good luck!

 

PCs assume that the hardware is working in most cases. There is a CRC check of data at the output port of the HD and there are varying degrees of checking on LAN copying but it is still possible to get undetected errors when doing that. RAM does not have even simple parity bit checks on typical PCs but proper servers will likely have ECC.

The most likely cause of no automatic verification when the archive is created is time. Most people would prefer to do the archive quickly and then do a check if desired.

The validation is not a "compare the data in the archive to the data on the HD". It opens the archive, reads it into RAM and reconstructs the 4000 checksums/GB and comparest them to the ones placed in the archive when it was created. It is a pretty stringent test and it checks the data as placed into RAM not just at the port of the HD.

Based on the above, anything that causes a failure of the checksum process could be at fault and this entails a significant portion of the PC. A bad sector on the disk or device holding the archive, bad RAM, bad disk cable (cause of my validation fault), bad CPU as found by at least one poster, bad motherboard, marginal power supply.

The vast amount of data read and written by TI at high-speed can show up problems in marginal hardware that are not seen in normal PC operation. Also, PCs that are overclocked or are being run with aggressive memory timings may fail while running TI but work OK for other work.

The validation should also be done using the TI rescue CD which is Linux and it may or may not have good drivers for your system. This is the environment that must run properly to restore the active partition even if you start the restore process in Windows. The rescue CD should be tested on a system before it is really needed and this is where a lot of TI users who have problems come unstuck. After you know the TI CD works on your PC you can put it away until you really need it unless you change your hardware. By far the best test of the TI CD is to restore your archive to a spare HD in case the restore fails. One of the first things TI does before restoring a partition is to delete the partion so a failure can be a big problem.

As always, testing that you can restore an archive of any type is not something peculiar to TI - this is proper practice for any backup program.

 

.
I did use the TI Rescue disk today and having discovered, a couple of weeks ago, that ATIH has a problem backing up the Vista SP2 partition I used the rescue disk then to avoid the problem and when it finishes backing up it announces that the operation was successful even though apparently it doesn't know!
.