Who DOES NOT have a problem with corrupt images...

[hoser_d]

when backing up to an USB external hard drive? This seems to be a problem for a lot of people. I am curious if any people do not have problems with corrupt images on USB external hard drives. Maybe, if there are people who do not have problems with corrupt images then we can better understand why many of us do have problems with corrupt images.

I use the latest build of TI and my EXT HD is a USB 2.0 160 GB WD.

[hgratt]

Be aware, I have never had to do a restore.

But, I use TI 7, Build 638. I have a Dell Dimension 8300, XP Pro, SP2, 80G HDD and an IBM R40 laptop, XP Pro, SP2.

I image to an external USB 2.0, Seagate 160G HDD.

I have yet to see a "corruption" and I verify each image I make (about 1 per week per machine) with the TI 7 check image tool.

Harvey

[hoser_d]

Have you recently gone back to verify some of your images? Or do you just verify the image when creating it? I don't have any problems creating images and TI says that the image was successfully created, but when I go back and verify the images at a later time, every single one on my EXT HD is corrupt.

[hgratt]

A good question. I am now in the process of re-verifying my Desktop images. Should take about an hour - I will post back with the results.

Harvey

[jimmytop]

After Acronis reports "Image Created Successfully" or whatever the exact words are, I always do a Check Image.

Since I've owned Acronis TI 8 (never owned a previous version), I've not yet had a Check Image report a corrupted image.

Always used with NTFS files system and PATA hard drives, FYI.

[mdburkey]

A couple of questions....

1) Are you backing up to the external USB drive from inside Windows or from the Acronis Boot CD? Or perhaps using TI for Linux?

2) Who made you external USB drive bay and what chipset does it use?

The reason I ask is that some USB external bays (especially many of the combo 1394/USB2 bays) use a Prolific IDE bridge chip which has many KNOWN issues with Linux (including data corruption) unless you have updated the firmware to the most recent version.

[earther]

I have been imaging to a 30GB external Quantum housed in a Metal Gear Box USB2 case. My system is WinME (getting a new box with XP this week), 20GB hard drive (less than 5GB used). I had never verified an image after creating it supposedly successfully. But after reading all the threads about corrupt images, I thought I'd give it a try. Well, it checked out perfectly without errors. I'm using TI8, build 786 and have been imaging from within Windows. Of course, I haven't tried to restore an image yet. In fact, I haven't yet even tried booting from CD or floppy!

Originally Posted by hoser_d

when backing up to an USB external hard drive? This seems to be a problem for a lot of people. I am curious if any people do not have problems with corrupt images on USB external hard drives. Maybe, if there are people who do not have problems with corrupt images then we can better understand why many of us do have problems with corrupt images.

I use the latest build of TI and my EXT HD is a USB 2.0 160 GB WD.

Hi,

I have NOT any problem backing up and/or restoring to/from my external USB-2 drive.

I run Windows 98 SE.
Latest TI version/build.
System: P3, 600 MHZ, Mobo Asus P2B, 512 MB RAM.
Sitecom USB-2/Firewire card in my system.
Maxtor OneTouch 80 GB external USB-2 drive, connected to that Sitecom card.

I always use the TI bootable floppies (write protected) to make backups and/or restore a backup.

I only use TI to make full backup images.

I hope this helps.

[hgratt]

Originally Posted by hoser_d

Have you recently gone back to verify some of your images? Or do you just verify the image when creating it? I don't have any problems creating images and TI says that the image was successfully created, but when I go back and verify the images at a later time, every single one on my EXT HD is corrupt.

I booted up my Dell desktop (DIM 8300) and attached the external HDD, etc.

I was able to sucessfully verify the last 3 disk images that I performed. I will do the same with my laptop and post back in a couple of hours.

Harvey

[Menorcaman]

Hello earther,

In fact, I haven't yet even tried booting from CD or floppy!

Please do yourself a favour right now and check that TI detects your external USB drive after booting from the rescue CD or floppy. You don't want a nasty surprise come the day TI is used to restore your system partition/drive!!

Regards

[earther]

Originally Posted by Menorcaman

Hello earther,

Please do yourself a favour right now and check that TI detects your external USB drive after booting from the rescue CD or floppy. You don't want a nasty surprise come the day TI is used to restore your system partition/drive!!

Regards

Yeah, I know. It's just that I don't feel totally comfortable in boot disk environments. And I would not be a happy camper if I screwed up my working system.

[hgratt]

Originally Posted by hgratt

I booted up my Dell desktop (DIM 8300) and attached the external HDD, etc.

I was able to sucessfully verify the last 3 disk images that I performed. I will do the same with my laptop and post back in a couple of hours.

Harvey

O.K, I sucessfully verified the last 2 images on the external HDD for my laptop. FWIW, all of my images are about 17GB in size.

I also verified (for both machines) that I could cold boot and read the Recovery CD via F12 and see my external drive. I went all the way to the point of committing to a restore before cancelling out. So, at least for my plain vanilla systems, TI seems to work.

Harvey

[brandis]

Hello I want to report my own bad experiance with corrupted images!!

I work in a small company with 6 PC all with W2000, we use TI since version 60 now 80 to make IMAGES of all PCs. We use one W2000 PC as "File server", so normaly we do image over our 100Mb LAN on the internal disk of our "server pc" (max. 80GB). As extension we have external disk,a MAXTOR PS5000XT with 250GB over USB2.0. On this drive I COPY many of the images that I do with TI to get free space on the 80GB drive!

After problems with one pc I wanted to install a image I did save one day before. I booted the pc (a compaq armada 500) with the acronis cd, get access to the server and tried to verify the image before doing it. After 1,5 hours I did get the message that the image is corrupted!!

then I did run the verify on the server pc direct (of course faster) but with the same bad result!
the image file hase 6 parts, one full image and 5 incremetal images. this different parts was created with different versions of TI (70 and 80) so I dont know if this has any bad affects.

To investigate this I did copy this images on our MAXTOR 250GB via USB2.0 and took it home hoping to find out what is going wrong!

I did connect the MAXTOR 250GB USB2.0 to my home pc, again the image is shown as corrupt.
I checked now more image files that I have on the MAXTOR 250GB AND THE MOST OF THEM SEEMS TO BE CORRUPT, now this is a lot of TROUBLE for me!! WHAT IS GOING ON HERE?

In the past I did use the MAXTOR 250GB to backup a my Acer Labtop directly with 1394 firewire(100Mbit because its a small 4 pin connector) and not with USB(because only USB1.1 is available) and all was working perfect.

So I did start testing.

I took a TI image created with my home pc and check the MD5
20041127_ath3200_c.tib file size 1,279,563,776
1. Original file on PC-Disk IDE 846535372F221134358E4E178C375ABA
2. Copy on other partition of IDE disk 846535372F221134358E4E178C375ABA
no problem when copying on same IDE disk
3. Copy on MAXTOR via USB 2.0 405D8DC1A88334EFD21CCD3827B128A
MD5 is wrong only with copy!!
4. Copy on MAXTOR via USB 2.0 846535372F221134358E4E178C375ABA
now I did copy again BUT on other folder, this time seems to me o.k.
Again calulating MD5 on file 4. !!! 7740570E1947598B2FF67B67556A222D
MD5 changes every time I did calculate it on the same file on the MAXTOR 250GB via USB2.0!! THATS REALLY SCARING

Now a some test with a different file
(300MB demo.ZIP) on MAXTOR 250GB over USB 2.0
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
MD5 is the same every time I check it so all o.k.

Now I copy this file from MAXTOR via USB on the IDE disk
Copy of (300MB demo ZIP) from Maxtor 250GB over USB 2.0 to IDE-Disk on PC
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
MD5 is the same every time I check it so all o.k.

Now I copy this file from IDE disk back to MAXTOR 250GB via USB2.0
Copy of (300MB demo ZIP) from IDE-PC-Disk to Maxtor 250GB over USB 2.0
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
C4DA88E1D585DF8BBC3269AEEF52B420
MD5 is the same every time I check it so all o.k.

Now I did copy a SMALL TI image from IDE to MAXTOR 250GB via USB
K:\Acer TM632LC\MeinBackup_D3.tib file size 282,465,792 to MAXTOR 250GB over USB2.0
01F3C99F94E5CCDC7B134847AD4733FA
01F3C99F94E5CCDC7B134847AD4733FA
01F3C99F94E5CCDC7B134847AD4733FA
01F3C99F94E5CCDC7B134847AD4733FA
MD5 is the same every time I check it so all o.k.

Now I did copy a BIG TI image from IDE to MAXTOR 250GB via USB
K:\20041127_ath3200_c.tib file size 1,279,563,776 on Maxtor 250GB over USB2.0
49F1F5B912E4E4C823E4113E5F37E394
65218506CBA239597D67359C7643811E
109CA9D210A73F11698FDBBE7B2A1D3E
MD5 changes every time I did calculate it on the same file on the MAXTOR 250GB via USB2.0!! To be clear I am checking the same file without doing anything between the checking!!

Now I did create a BIG ZIP-file to check if this happens only with TI files!
K:\Game-Demos\test.zip filesize 1,165,303,696 on MAXTOR 250GB over USB2.0
00AF4DEC1E51396281CF129E046B49A9
1F28BF8FE82657CB4DBC352CCAFCC0EB
A9E76822FC3B66C8F7BFDB48A0B465DC
7E3F9FB5730C5A3CEA56206ABAE6824E
MD5 changes every time I did calculate it on the same file on the MAXTOR 250GB via USB2.0!! To be clear I am checking the same file without doing anything between the checking!!

Now I did copy this BIG ZIP-file to the IDE disk
E:\test.zip filesize 1,165,303,696 on PC-DISK-IDE
DCA853AA02EE40E3B3805CE09A05CB57
DCA853AA02EE40E3B3805CE09A05CB57
DCA853AA02EE40E3B3805CE09A05CB57
MD5 is the same every time I check it so all o.k.

now I did copy the same file back from IDE-disk to Maxtor via USB2.0 on different folder
K:\Game-Demos\test.zip filesize 1,165,303,696 on MAXTOR 250GB over USB2.0-BACK
DD14F7F16E1D6A298EBF1959E39954BB
18EFDCC471238CAB3A97464FFDCE5856
8CFBF6EE073D4525244E3780B96B50E0
MD5 changes every time I did calculate it on the same file on the MAXTOR 250GB via USB2.0!! To be clear I am checking the same file without doing anything between the checking!!

So what does this all mean??
For the moment it seems that I and many other users are having BIG PROBLEMS using USB2.0. The different results of MD5 on the same file can mean only that we have BIG PROBLEMS READING THE DATA CORRECTLY FROM THE EXTERNAL DISK. THERE MUST BE ERRORS IN THIS READING OPERATION! Of cause we can also have errors with writing/copying the files, but if the error was only there we would have differtent MD5 results compared with the original file and not every time calculating the SAME file!
As my test show this seems to happen ON BIG FILES, the 300MB file seems to have no problems, the 1.2GB file has allways problems!

I did not check if this problem is the same if you copy over LAN (100Mb) or 1394 FIREWIRE, but remembering the coruption problem I did have with the CompaqArmada500 I think that this is POSSIBLE an it should be checked and verified by other members!!! I personally will do the same tests with the MAXTOR 250GB using its Firewire 1394 port as soone as I get a cable (I did forget to take it from the office).

For now I did a small check using the Acer Labtop conecting the Maxtor 250GB with 1394 firewire (100Mb)
K:\20041127_ath3200_c.tib file size 1,279,563,776 on Maxtor 250GB via 1394
The MD5 is every time THE SAME!!!
This comparison is a bit unfare because the Acer is far slower the my home PC (Acer P4/1600Mhz , Home-PC ATHLON64/3200) and the 1394 is not full speed (400Mb) but only 100Mbit/s.
But it give us a hint that perhaps the READ-ERRORS acure more often if the USB2.0 is driven to its limits. Or USB2.0 hase problems if the file size exeeds a limit, ( remember that I didnt have problems with the 300MB files).

I dont know what it is but I hope ACRONIS will help us to resolve this matter! Perhaps ACRONIS must add a double check while sending/receiving the data to check that the port(USB or 1394 or LAN100Mb) is not sending/receiving rubbish, or you have to add a SLOWDOWN BRAKE not to drive the ports to the limits, or some other trick.

If the hardware hase really such problems then it seems to me we will not get many support from the manufactorers. The pc MFG will say its the error of the USB-card on your PC, the USB-card MFG will say its the external drive error or the chipset of the IDE-USB bridge, the external drive MFG will say its the PC-USB card error, all will say perhaps its a defect cable.......... and we will be the stupid users....

Stefanos Albanidis
sa@tronicpool.com
Tronicpool GmbH
Germany

[hoser_d]

Well, my PC is a PIII 1 GHz, so my computer was not initially equiped with USB 2.0 ports (not sure when they first started appearing). Not until recently have I purchased a USB 2.0 upgrade card so now my external WD 160 Gig HD is connected to the USB 2.0 card, however; all the images that I had on my EXT HD were created using a built in USB 1.1 port, (very slow, hence the card upgrade ) so I'm not sure if the USB 2.0 theory would hold up.

[Ed Y]

I've tried creating the image on an external USB drive (Teac with a Western Digital internal HD) using both USB 1.1 and 2.0. Same corrupt problem. Funny thing with me is that back in Oct/Nov with the release of TI 8.0 I was using then, the corrupt problem only happened about every 3rd time and always would correct itself if I kept doing the runs over again. Now with the latest version, the image is always corrupt.

[jmk94903]

Stefanos,

You have done an amazing piece of good work. The changing MD5 for the same file is amazing and seems to explain so much.

One idea for using USB drives would be to always split the backup files at 747 or 896 MB to keep the largest file below 1,000MB. I picked these because they are 1/6 and 1/5 the capacity of a 4.7GB DVD disk. Of course, split at 700MB if you use CD-R disks. That means more files but no corruption if your work checks out with others.

I'll run MD5 checks on several files on two USB 2.0 disks that I have to see if I can reproduce your variable MD5s.

John

[Acronis Support]

Hello all,

Thank you for your readiness to help us in the investigation of the problem. We do appreciate the work you do. All the results of your invetigtions are sent to our Development Team and they take these results into account.

Thank you.
--
Ilya Toytman

[brandis]

Originally Posted by Ed Y

I've tried creating the image on an external USB drive (Teac with a Western Digital internal HD) using both USB 1.1 and 2.0. Same corrupt problem. Funny thing with me is that back in Oct/Nov with the release of TI 8.0 I was using then, the corrupt problem only happened about every 3rd time and always would correct itself if I kept doing the runs over again. Now with the latest version, the image is always corrupt.

Hello Ed
forget the tests with TI.
Do the simplest test!!
Creat a big file (or better try differt file sizes beginning with small like 100MB) on your local disk. Check the MD5 3 times!
Copy this over USB 1.1 to your external disk. Check the MD5 3 times!!
Copy this over USB 2.0 to your external disk. Check the MD5 3 times!!

If copying the file and calculating MD5 gives you different results EVERY IMAGE WILL BE CORRUPT!!

Stefanos Albanidis

[Ed Y]

Originally Posted by brandis

Hello Ed
forget the tests with TI.
Do the simplest test!!
Creat a big file (or better try differt file sizes beginning with small like 100MB) on your local disk. Check the MD5 3 times!
Copy this over USB 1.1 to your external disk. Check the MD5 3 times!!
Copy this over USB 2.0 to your external disk. Check the MD5 3 times!!

If copying the file and calculating MD5 gives you different results EVERY IMAGE WILL BE CORRUPT!!

Stefanos Albanidis

That's a good idea. Haven't tried that yet. I took the external HD to a small local PC shop to get them to test it on a couple of their machines since I don't have access to any others. I'll do those tests when I get it back, hopefully tomorrow or Tuesday.

[Tsu]

Originally Posted by brandis

Hello Ed
forget the tests with TI.
Do the simplest test!!
Creat a big file (or better try differt file sizes beginning with small like 100MB) on your local disk. Check the MD5 3 times!
Copy this over USB 1.1 to your external disk. Check the MD5 3 times!!
Copy this over USB 2.0 to your external disk. Check the MD5 3 times!!

If copying the file and calculating MD5 gives you different results EVERY IMAGE WILL BE CORRUPT!!

Stefanos Albanidis

Large File Corruption is an ugly situation and Google will give you a preview of what you are up against http://www.google.ca/search?q=large+...=&start=0&sa=N

While you puzzle out the unique solution for YOUR environment, your first priority, IMHO, if you are running mission critical backups/images, is to see if any smaller file size will copy okay and then start Acronis on all those systems with that setting immediately. Then persue your hardware/BIOS/USB permanent solution

Remember, this is not an Acronis problem as they are just using the your O/S and hardware to do reads and writes. Luckily for me I have only encountered this once before on a single system and was accidentally discovered when copying large MSI files to an archive disk and then later trying to use them to reinstall applications. This wasn't even a USB disk.

WINZIP and other archivers are useful tools for testing as you can easily create a file size of your choice using multiple copies of a single 100mb file, then, try and extract all from this archive with no CRC errors.

Good luck to all

[hgratt]

FWIW, as I posted above I use TI 7, B638 and have not seen the corrupption isssue. I image to a Seagate 160GB, external USB 2.0 drive which is formatted with FAT32.

My images are around 17GB in size and thus require 4 or 5 files. The largest ones are 4GB in size.

If this is an OS issue (not a TI problem), then I wonder if this problem is encountered with other imaging programs (i.e., Ghost 9, etc). Would there be a way to post this issue in a more general "storage" newsgroup?

O.K, I read my share of the 800,000+ hits on the link supplied by Tsu. The 5 or so threads I chased down seem to imply that the corruption issue is a hardware problem. This is going to be an interesting topic of discussion!

Harvey

[CooL3]

Lucky me, not seen it yet.

Just for you to know, if you're using an external USB2 HDD (like me) the drive should be set to CS (Cable Select).
This because USB chips are not very smart.

Hope this helps.

[jimmytop]

Originally Posted by CooL3

Lucky me, not seen it yet.

Just for you to know, if you're using an external USB2 HDD (like me) the drive should be set to CS (Cable Select).
This because USB chips are not very smart.

Hope this helps.

Anyone else try setting their drive to cable select as Cool3 suggests There must be some reason why he hasn't had this issue that everyone else has....

[hgratt]

Originally Posted by CooL3

Lucky me, not seen it yet.

Just for you to know, if you're using an external USB2 HDD (like me) the drive should be set to CS (Cable Select).
This because USB chips are not very smart.

Hope this helps.

Did you buy your own enclosure for your drive? Is this where the CS jumpers are located?

My Seagate External USB drive was a complete package and I see no place to change jumper settings.

[brandis]

Originally Posted by CooL3

Lucky me, not seen it yet.

Just for you to know, if you're using an external USB2 HDD (like me) the drive should be set to CS (Cable Select).
This because USB chips are not very smart.

Hope this helps.

What do you mean with "the drive should be set to CS (Cable Select)" ?