Introducing AX64 Time Machine - hybrid imaging/snapshot software

I'm kind of in the Pete camp on this. As a pure IMAGER, there are many others out there that do a more than adequate job at imaging (Macrium, Shadow Protect <$$$$>, IFW, etc.) but none of those have any good snapshot capability, which is the main reason I'm even looking at this tool. If its imaging capability meets my specifications, without scaring me that is, then all the better... but it's the snapshotting that I'm most interested in.

I'm a long time Rollback RX user but have left the fold as the product has moved into the v10 iteration... it's just too damn flakey and very system configuration dependent. The main difference between Pete and myself is... Pete wants good, solid imaging capability, and if that solution happens to develop into a good solid snapshotter... great, he can always use that capability. I want good, solid snapshotting and if that tool develops into a goto imager, all the better. Where Pete is providing goto backup with Macrium, I am doing the same with IFW. If Time Machine ever gets to the point of "perfect" imaging then maybe it will become my 1-stop shopping item for both requirements.

At the moment I am very happy (barring fixing some BETA anomalies... like a REAL COLD RESTORE!!!) with Time Machine even for my imaging... I know its current liabilities and am willing to put up with them (along with IFW as my lifeline).

 

Don't forget Shadow Defender is incompatible too.

 

Your "dear Froggie!?!?" <awwwwwwww...>

Yes on the PageFile. I'm not sure about your current issue is but Track 0 is an absolute requirement (Time Machine Devs... pls hear this!!!). Many applications use that area when using the original "extensible" BiOS options that were available with MBR systems before EFI came along... BOOTit Bare Metal is surely one of them, there are others.

I'm very interested in hearing about the anomaly you continue to see... what's actually happening?

 

I would guess this is the case with almost all "light virtualization" (LV) apps available... Shadow Defender, Returnil, Deep Freeze, there are lots of them. Any application that's trying to track the system for whatever reason will have a problem with these. Since Shadow Protect uses a tracking mechanism also, I'd have to guess it doesn't like LV apps either.

When apps start using these esoteric system options (low-level NATIVE SYSTEM APIs, fancy personal drivers <Acronis is the KING here>, etc.), all kinds of things begin to happen. You start entering into the world of the wild, wild west... just like the MBR and Track 0 have been since the beginning. You just never know how you're going to affect any other task on the system when you do things like this. And the "Doer" can't possibly figure out what effect they'll have until they do it.

Just look at Rollback (an MBR example) and tons of other tasks when the EXTENSIBLE FIRMWARE INTERFACE (EFI) hit the mainstream scene... kaboom! EFI has been discussed for many years with Intel leading the way with specifications (and use) but not until U(nified)EFI firmware appeared on mainboards did the blowup occur (and it's still going on ). This type of stuff is inevitable (get ready Pete ) and always occurs at evolution points along the way. Knowledge is the only way to keep it at bey, and if that can't be gained during your travels for whatever reason... buy a DELL and let somebody else try and worry about it. It's definitely getting waaaaay more complicated.

 

Would someone kindly link the (old) version download list? I thought I'd saved it but I can't find it. Thank you!

 

Are you looking for http://ax64.com/beta/ ?

 

That works and thank you!

 

Thanks for your comments, I'll report those issues and see what they say.

 

Pete, thanks a bunch for all that testing... a lot of work but good insight learned.

It's as I expected. What PD does during BOOT defrag is unseen by the next incremental attempt (PD bypassed TM's tracking mechanism), but fully seen during a BASELINE run ('cause it's saving everything at that point). Like I surmise, I don't think this problem will be fixable between TM and PD unless TM can figure out how to look down deep to see what's been changed.

Your last sentence above... did you really mean WOULD or do you still mean WOULDN'T?

 

Finally I had some time to test the new v.2 Beta. I first restored a WinXP image to my laptop (older MSI notebook, AMD dual core CPU, 2GB RAM, 160 GB HDD). Installation went fine.

Creation of a recovery USB stick was slow, but it did work, and the laptop could boot from it. My HDD had 3 partitions, I selected them all and did several backups, all uneventful. The backup browser showed all the backups, but mounting a backup failed each and every time. Not too catastrophic because recovering files directly from the backup browser worked fine.

Several hot restores also worked. Of course I could not install the bootup recovery console, and warm restores could not be selected. But this was the expected behavior for WinXP.

My next test was doing restores from the USB recovery stick. To my surprise these were always true cold restores (as opposed to what Froggie and Peter2150 experienced). A WinXP thing probably.

The last test was imitating a total HDD failure, I wiped out the MBR and partition table of the HDD and then tried to restore my system using the recovery USB stick. This was a complete failure and makes AX64 v.2 in its current state completely unusable.

There is no option to restore all partitions at once, you must restore every partition separately. When I selected the system partition c: a window popped up informing me that the original partition could not be found and giving me a choice of the physical drive to restore to. After selecting the correct HDD I got the error message: "Cannot create partition structure on the disk". The same thing happened when I selected one of the other 2 partitions that were backed up.


After this total failure I am not sure if I even should take the time to make more tests under Win 7. A backup software has to get me back up on my feet after a total HDD disaster, and all traditional imagers I have used can do this. I guess I will wait for the next AX64 version...


Cheers
manolito

 

Manolito, at this point in TM's evolution it's still not a DISK recovery tool, it's a partition recovery tool. In its first iteration, it only saved 1-partition and restored only 1-partition. This 2nd iteration just adds additional partitions to the save set... but NOT the disk. Other imagers do offer multi-partition and/or FULL disk saves but TM does not. Since it was designed out of the box to be more of a snapshotting tool than an imager, that seems to be the plane that it's riding on at the moment and seems to be a "Partition Protector" more than anything else.

 

Hokie Dokie fellas, so if TM v2 post-beta fixes key bugs, but stays essentially the same in regard to not having a true cold restore and the issues with PD, then would the following be solid, theoretically viable situational protocols?

1. In the case of system failure or other catastrophic PC situation:
A. Use a "monthly" Macrium Reflect backup to restore the system;
B. Then use TM to snap to the most current state.

2. In the case of just human error silliness (not mine, of course! , just use TM's warm or hot restore.

And if that makes sense, am I missing a simpler option?
And if the above makes sense, can I use all aspects of PD without creating a mess?

Thanks,
Chamlin

 

Chamlin, I agree with Pete's statement above. But HEED HIS WARNING about using PD's BOOT Time Optimization. If you do, you need to start a new incremental chain in TM with the first being a new BASELINE image following the PD off-line process... otherwise you're looking for trouble.

 

On behalf of anyone else who feels this way but hasn't yet expressed it, just want to thank Pete and Froggie for all the testing, guidance and wisdom being shared with us here. Makes a world of difference in our computing lives!

And to Waj, Isso, and the rest of the team, I hope you know that all of the recent conversation about v2 has to do with all of us wanting TM to be a future classic program and that we want you/it very much to succeed, yes? We're behind you and support generating something great here for ourselves and for you.

Where are you guys at / what is your vision in terms of whether or not you currently intend to develop true cold restore capacity, be it in v2, v3? Once we know this, we can all figure out how to best proceed with our PC survival protocols.

Thanks,
Chamlin

 

@Froggie
This is from Issos very first post of this thread. The key word for me is "Can be used as backup in case of disaster". For AX64 v.1 this is still true. It can restore the system partition even to a totally wiped HDD leaving the user with a working and bootable system. Of course the user must then repartition the drive and restore his other partitions. A little annoying, but doable.

If v.2 does not follow this basic concept any more (traditional imaging based backupper with the additional feature to allow fast hot or warm restores) then it will not be useful for me any longer.


Cheers
manolito

 

I agree... hats off to all the active testers here!

To the AX64 team:
I have been reading this thread since January of 2013 (yes, all 300+ pages) but AX64 has never been quite complete for me. The multiple partitions solution (which I think is pretty critical) was going to finally get me on board.

However, based on the cold restore issues, I just bought another copy of Rollback during yesterdays sale. I know the risks of Rollback (as well as GoBack, CTM, RestoreIT, etc.)... and can manage them. Your solution has to be bulletproof (for me).

I am just writing to let Isso & Waj know that although only a small group post, there is a lot of visibility to this thread. If you make this work (before one of your competitors does), you will have a very marketable product.

Regards,
SteveG

 

Hi Steven! I guess I'm just a bit surprised at your statement above. Some of the last things I would ever refer to as "bulletproof" would be Rollback RX or CTM, among others.

Clearly your mentioned Rollback RX can do almost nothing for you as far as hardware protection is concerned... what app do you use for your system or disk backup?

 

Morning Manolito!

As you can see from Post #7871 from Wajamus, they recognize the current COLD RESTORE shortcoming in the RECOVERY MEDIA of TM v2 and plan to offer that before TM v2 comes out of BETA in this cycle. It should be as effective as AXTM v1 barring the Perfect Disk BOOT Time defrag anomaly recently tested by Pete... please keep that in mind.

 

And you think Rollback will be bullet proof for you? Good luck.

 

To be fair, StevenG seems to be saying that he knows Rollback isn't bulletproof, but can manage the issues. Presumably for some reason he doesn't feel able to manage TM's issues.

Personally I love AX64 v1 and v2, and just always ensure I have a secondary image made with some other program. But then that's probably the sensible thing to do regardless of what your primary imaging software happens to be. I imagine there's a place for using Rollback accompanied by an imaging program too. I wouldn't expect these snapshot programs to excel in cold image restoring - it's not their focus.

Incidentally, as far as I can tell, I caused v2 to do a cold restore by first instigating a hot restore on the D partition, followed by a warm (though apparently actually cold) restore on the C partition. That's what I think happened anyway - I can't say for sure as I cut the power to my pc due to a bios-related pause partway-through the restore process.