Corrupt tib files

Discussion in 'Acronis True Image Product Line' started by Al Fischer, Feb 3, 2008.

Thread Status:
Not open for further replies.
  1. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    TI11 Build 8053
    AMD w 1 gig and WD Sata drives.

    I recently had need to restore my system. Had the tib on another system on the LAN. From rescue disk I selected the proper tib file and started a restor. Got an immediate 'corrupt' message. Went to the other system and copied the tib to a USB drive. Returned to system in question. Tried to restore. Immediate 'corrupt' message. Took HD to the system containing the tib and it restored without ussue. Back to orignal system with HD, installed it. Booted fine.

    1. The tib is not actually corrupted, as it could restore to the same HD from the taken system under question.
    2. Something is apparently a bit strange with my system in question.
    3. No other problems reading files, burning DVDs, Virus scan done, never a boot problem and networking never has had a problem. Can copy internal HD to internal HD as well as to external USB or eSATA drive.
    4. CHKDSK runs error free.
    5. MEMCHK 86 rens error free.
    6. Basically system works fine except for falsely declaring a tib as corrupt.
    7. I KNOW that it worked perfectly at one time both with TI11 as well as the rebadged version from Apricorn.
    8. The only reason for restoring was that I had hosed some software and this was the path of least resistance.

    o_Oo_Oo_Oo_O Where do I start troubleshooting this problemo_Oo_O
     
  2. Peter Mac

    Peter Mac Registered Member

    Joined:
    Oct 7, 2005
    Posts:
    73
    I copied my TI 9 image to a USB drive and then tried to restore from the USB and it also said corrupt.

    My experience with Acronis is that you can't move the .tib files or they will get flagged a corrupt. This is poor programming, but that's how it is.

    They seem to restore when left in the same HD spot and copied back to the same machine. I no longer trust the reliability of ATI images. I am looking for a more reliable image producer.
     
  3. blockie

    blockie Registered Member

    Joined:
    Aug 26, 2007
    Posts:
    73
    For Peter MAC.
    When you find a good one please let me know at blockie@sbcglobal.net.
    I updated to TI11 and find all kinds of errors. The scheduler is unreliable on back up but OF for validates. Every validate produces a "corrupt file". Don't know if it is The validate process or if every single one of my archives are corrupted.

    Bill
     
  4. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    I have not found it to be unreliable. I have created and restored both from the same system and from another system. Files on any HD or CD/DVD. I have cloned many times with no problems. And I mean MANY. Even to creating an image of a new laptop before it was ever booted and capturing the factory supplied, unconfigured image including the hidden service partition. It restored to a virgen drive without issued. (and booted). This was done both to another HD as one file and to several DVDs. Hey, testing before loading the software is the only way to fly. Besides most times a decent restore process is not supplied from manufacturer.

    The problem I am encountering is new and 100% consistent. Does not matter where the tib was stored. It fails to restore if done on this system. Prior to now it worked. I suspect conflict with other software but have no idea which is the guilty party. Or where to start.

    I would REALLY prefer to not have to build a new image as my IMAGE is about 30gig. Lots of software. (mostly unused!!)
     
  5. h2ouup2

    h2ouup2 Registered Member

    Joined:
    Jan 31, 2005
    Posts:
    25
    Do you have KAV 6,7, or 8 beta by any chance?


    I read where it sometimes the ObjectID's not only have problems with ChkDsk but ATI as well, but I haven't looked into it much at all, so I could be way off.

    I don't use KAV anymore, and TI 11 has been rock solid. No corrupt images, and restores are fast, but YMMV.
     
  6. blockie

    blockie Registered Member

    Joined:
    Aug 26, 2007
    Posts:
    73
    I don't know what KAV is.
     
  7. IBK

    IBK AV Expert

    Joined:
    Dec 22, 2003
    Posts:
    1,886
    Location:
    Innsbruck (Austria)
    KAV = Kaspersky Anti-Virus
     
  8. TheQuest

    TheQuest Registered Member

    Joined:
    Jun 9, 2003
    Posts:
    2,304
    Location:
    Kent. UK by the sea
    Hi, Peter Mac

    When you move or copy anything including a .tib file it has nothing to do with Acronis, it is explorer [part of the OS] that is in "charge," and faults and errors can usually be caused by memory.

    Take Care,
    TheQuest :cool:
     
  9. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    Absolutely, also writes onto a bad spot on the disk or a spot that goes bad later. When files are copied there is no read after writing check so you don't know it is bad until you try to read the file. TIB files seem more problematic just because they are so big and few applications have such a rigorous error checking as TI with one checksum every 256K bytes.

    Rather than get anal about the word corrupt, re-word the message to say that "for some reason TI can't read the file properly". This opens up your hardware and when restoring the active partition the TI Linux recovery environment to be potential candidates for the cause. If an archive validates in Windows but not in the recovery environment, either bootCD or when Windows requests a reboot to recover, then the fault most likely is with poor Linux drivers or a related problem. Hardware can still be a problem due to different RAM mapping and possible different access methods to the disk but is less likely.
     
  10. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    I don't have KAV. I use AVG Free. This worked before so I don't suspect AVG but I tried turning it off and no difference. So far as HD failures this problem is with IDE, SATA, eSATA, USB and Firewire attached drives all of which run chkdsk clean. (mostly recent WD drives) I just tried and it actually failed half way through creating an image. Have run MEMCHK 86 for 24 hrs and no failures. Also 'Stress Test' (I forget whos but no probs reported. Can copy large folders with no problems reported. I think burning DVDs probably is a good stressor. Where is it docummented that XP does not do a read-back check when copying? I was not aware of this. Is there a 'copy utility' that does a read-back check?

    Believe me, I am open to any and all suggestions as I really need this machine backed up securely.
     
  11. shieber

    shieber Registered Member

    Joined:
    Oct 27, 2004
    Posts:
    3,710
    More likely, ATI has probs read/writing to some usb setups. Copying files isn't an issue, it's ATI ability to read/write on diff hardware setups. You can make valid tib on an interanl drive, copy it to usb, then ati can't read it properly on the usb and reports it invalid. Try copying back to the disk where it originally registered as valid and see if it isn't still the same content.

    If not, then your pc has problem with disk read/writes and it's not particularly an ATI prob.

    If so, then ATI has prob prperly read/writing fromthe usb drive.

    There are rarer situations where intermitient memory probs or bad memory in a very high range not normally used except when dealing with very large files can cause file read/write probs. You cant est for these issues by running memtest for a dozen or so hours.


     
  12. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    Try Memtest86+, V1.7 from www.memtest.org if you want to try another one.

    You can verify file transfers by running a checksum calculator like:
    http://www.irnis.net/soft/xcsc/

    Run it on your file and record the checksum (you don't need to do all 3, the MD5 is a good one). Record the value. Transfer the file and run it again, the values of the checksum must agree. Note that a memory problem could cause this to fail as well.

    Transfering a large folder of various files is not the same thing as transferring a huge file in the GB range. Some USB chipsets particularly choke on very large files.

    What is your motherboard?
     
  13. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    It was MEMTEST86 that I ran for 12 hours. (when I said MEMCHK I meant MEMTEST!)

    I am on my way to download and try the checksum calculator.

    I am wondering about MEMTEST86 ability to test ALL of memory. It appears to not test the first 104K. However, if I had bad memory I would expect to be having some sort of problem with other functions.

    I have an ASUS A7N8X-E Athlon XP3200 1 gig corsair memory. WD36gig Raptor C:drive, 4 SATA 160 WD internal, 4 eSata WD320 externally, and several USB 2.0 drives.

    I tried booting directly from the TI CD and it failed during the image creation. Did a POR. That eliminates any SOFTWARE on the system and says I got a HW PROB! I have allready swapped out the MB and MEM as I have spares.

    What I don't have is a spare processor.

    Think I may remove all exept a single USB CD and 2 drives and try it.

    Please keep in mind that this DID WORK. And it worked VERY well. I also tried my old copy of the Apricorn version and got a failure. Apricorn worked for the last 2 years so the basic system config is compatible (not like the cases seen in Dell that I saw on this forum)

    GREMLINS??
     
  14. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    I think some processors may use the first bit of memory. I know my Intel PC won't do anything if low memory is bad.

    You seem to have swapped out the critical components so this is strange but you say it did work before and now it doesn't so something changed!

    The Linux boot CD version rules out any Windows/application problems because the disk is static when it is being used.
    Can you validate in Windows?
    Any chance overclocking enabled or memory settings are too aggressive? Pushing things can work normally but can cause TI problems.
    What about power supply voltages - any marginal such that high processor use (more power) and lots of disk activity causes the voltage to sag - you also seem to have a lot of disks on your system? Given your swapping other components I would certainly consider this.
     
  15. tuffshed

    tuffshed Registered Member

    Joined:
    Apr 19, 2007
    Posts:
    19
    Also have similar corrupt tib problems...help

    I own TI 10. I've had corrupt tib frequently, about 30-50% of the time. It happens on different external hard drives and different computers. I just figured that TI 10 was finicky. I finally came to this forum to see what's up.

    Here is my usual scenario. I boot to safe mode on the TI recovery CD which I build on the target computer beforehand. I do a complete MyComputer backup (takes hours as I have about 100Gb on disk), then I immediately run the check archive and almost half the time it blows immediately with a corrupt archive error message. I reboot and try the check archive again and if it failed originally it will always fail. I then erase the tibs, try a backup again and sometimes it works, sometimes not. I just got in the mode of repeating backup/check archive till I get a good one, but now I'm getting sick of this nonsense.

    What could be wrong?
     
  16. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    You really should have put this in a new thread and maybe a moderator will set it up.

    Why do you boot to safe mode on the CD to do the backup?
    TI has 3 modes (I'm going to talk about imaging backups):
    Windows - for creating the archive and for restoring any archive that doesn't contain the active partition, typically C.
    Full - Linux recovery environment, supports USB, network, Firewire.
    Safe - Some DOS environment. Doesn't support USB, network, Firewire but will do so on some modern motherboards. Typically very slow. It only is Safe in that it is a minimal system useful if archive in on an internal HD mostly.

    100GB of data should be imaged in less than 100 minutes on a reasonable system using Windows and maybe 1-2X longer if using the TI FULL version on the CD.

    Since it works about half the time, I'd suspect hardware problems first since they can be easy to determine - maybe.

    If the PC is overclocked then revert to normal settings.

    Memory settings should be the normal ones for the RAM, not aggressive timings.

    Download Memtest86+ version 1.7 from www.memtest.org note that it is 86+ version 1.7
    Let it run for a few hours, preferably overnight.

    Run chkdsk X: /r on all of your partitions including where you are storing the archive. Substitute the drive letter of the partition being tested for X. A reboot will be required for C.

    If this shows no errors, make an archive in Windows and validate it in Windows. If it works, try it a few more times. The Windows drivers normally provide the best performance compatibility with the PC hardware but they can't be used to restore the active partition unless you build a BartPE disk.
     
  17. tuffshed

    tuffshed Registered Member

    Joined:
    Apr 19, 2007
    Posts:
    19
    Seek,

    Thanks for the response. HW problems, I'm doubting this since I run TI 10 on both my laptop and my tower using one of 3 USB hard drives that I have (I rotate). I get corrupt tibs on either computer and I know I've gotten corrupt tibs on at least 2 of my USB hard drives.

    I didn't know the real differences between Windows, Full and Safe, but I need to use either the Full or Safe as I am always backing up my active partition (C: ).

    I am confused about Full and Safe though, you say it doesn't have USB support, maybe I dreamed it but I've always been able to see my USB drives in Safe mode, and I assumed Safe meant safe, i.e. using standard non fancy DOS drivers that have been working forever.

    I can believe if Safe doesn't fully support some more modern USB drives, and I'll be trying my backups with the Full to check this out. Do you think that this might be the problem, good enough USB support to read folder and file info and do writes but not good enough to do reads?

    As a HOME user, this doesn't give me much confidence in the product, that it appears to work but in fact hasn't. If Safe has problems they should have probably stated that more clearly in the instructions. In any event, all I really ever wanted to do was consistent safe backups. I'll try Full mode and see how things go.
     
  18. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    You can backup the active partition in Windows with TI. It has the smarts to back it up while Windows is running. I would do that and validate it using Windows. If it works then there is a good chance your system has problems with the Safe Mode stuff on the CD. Admittedly, the problem usually is with the FULL mode Linux stuff.

    Apparently some motherboards do allow TI Safe mode to run with USB drives via USB support in the BIOS it is said. I never use it.

    The typical way to run TI is to create archives in Windows and Restore starting the restore in Windows (assuming Windows is still operational) and letting it boot up the Full version (Linux) when it requests a reboot. If the active partition is not being restored then it will do the restore in Windows. If Windows is broken then the rescue CD is booted up and Full mode selected. I also have to admit I am surprised to find out how many people appear to use the Safe Mode. This became apparent from several posts when Acronis stopped automatically including it in the download with version 11.

    TI's Safe definition is similar to Microsoft's - it means you are getting the bare bones hardware support in case the other stuff is problematic, usually because of a driver or configuration error.

    Since you are having trouble with multiple PCs and USB drives I would agree that hardware problems are less likely but not impossible. Remember that TI flogs the disk system, processor and RAM a lot harder than most things you run. It is including and checking 400,000 checksums for your 100GB archive which contains roughly 800,000,000,000 bits - just 1 bad bit and the archive is declared corrupt!
     
  19. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    Well, looks like we beat my system into submission!!

    There were multiple problems. Looks like this started
    several months ago with failing BYTEC USB HD enclosure.
    It had read problem. Proved it by transferring and
    image over lan to USB drive and TII failed verification
    randomly. Ran the CHECKSUM (thanks for the link to a
    FINE piece of software) and it failed. Put this HD in
    IDE port and it always ran a good checksum. Many of the
    images I had created were apparently bad. Enclosure is
    in the trash (er, I mean 'recycle!)

    Now to the RAM problem. 1 gig dual channel of Corsair
    XMS. I replaced with 512 of Corsair Dual Channed XMS.
    Both sets are matched pairs and the RAM is certified
    for the MB. All was fine. Verification with the 512
    always ran good. Put the 1 gig pair back in.
    Verification immediately failed. 512 back in ran good.
    Put in 1 stick of the 1 gig pair. Ran fine. Replaced
    with the other stick of 1 gig, Failed always (now in
    single channel)
    First stick back in. No failures. Moved this questioned
    stick to each of 3 slots. Always fails in slot A2.
    Frequently fails in A1 or B1.
    So I have a very marginal stick.

    Now, Corsair recommends timing of 6-3-3-2.5 and 2.7 vDimm.
    The BIOS default is 8-3-3-3 and 2.6 vDimm
    Thats probably agressive even for Corsair. This system
    is not overclocked. I go for reliability not blazing
    speed. Figure if you want more speed get a faster CPU!

    HOWEVER it ran fine until several months ago. Should
    work. RMA TIME (Lifetime Warrantee is wonderful and
    Corsair does stand behind their product) If TII cannot
    handle memory timings that MEMTEST and XP can handle it
    would seem there may be a problem. Of course NO software
    can handle a BYTEC enclosure when it faile to read! (no
    more BYTEC for me. this is the 3rd bad one)

    So I continue to have faith in TII. I have used it and
    another version (rebadged) MANY times both to image and
    clone. This is first real problem in 4 years. It works
    and works well. IF your hardware is good.

    I do think it adviseable to VERIFY the IMAGE
    IMMEDIATELY after BACKING up. Better to discover a
    problem BEFORE you need to restore. Kinda like a fire
    drill? or a parachute test?

    Gentlemen, thank you for the assistance. This one was
    driving me NUTZ! (a short trip)
     
  20. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751
    Glad you found the problems! Having more than one problem certainly makes it more challenging.

    Your default RAM timings certainly aren't aggressive - they are typical of just about any old RAM. There should be no problem with TI working with memory timings that are in spec for the memory being used. However, there is no doubt that TI does give it a workout with such large, sustained data transfers being done as quickly as possible.
     
  21. Al Fischer

    Al Fischer Registered Member

    Joined:
    Feb 3, 2008
    Posts:
    21
    Another thought...

    What might you consider for more 'conservative' settings? I believe that Corsair suggests settings for their mdules that optimize speed. Seems that a lot of overclockers are around there. I understand that settings will affect benchmarks, but I couldn't care less about a benchmark. Only care about real world apparent speed with total reliability.. A single failure situation like I have have here does not make for 'speed'! I would rather go a BIT slower with 100% reliability than have a single failure.

    I run video rendering apps, photoshop, (with LARGE Tiffs)(30-50 MB), some audio as well as MS Office stuff.

    AND one of these days this system will be replaced with at least a dual core and 4GB w/raid 5. NOT this year. (money? yep)
     
  22. seekforever

    seekforever Registered Member

    Joined:
    Oct 31, 2005
    Posts:
    4,751

    IMO, you have the correct perspective. What's the point of running faster if the machine hangs and you have to reboot 3times a day or you have to do the task over because it hung - not really faster in the total picture. Same with benchmarks, you aren't running benchmarks to do the job you are running your app. So if the job is modifying a photo then the only benchmark that is applicable is timing the modify task(s) with a stopwatch; most people can't tell time changes unless they are about 10% or more.

    Memory timings are expressed in Wait states because the data can't be presented fast enough from the memory chips. In simple terms, wait states are inserted to slow things down until the memory gets its act together .

    The smaller the numbers given, the fewer wait states that are needed so the system runs faster. 2.5-3-3-6 is faster than 3-3-3-8. I think the first number, CL, is usually the dominant one regarding memory access but I am not sure - it seems to be the one talked about the most.

    I also have Corsair memory and I was surprised that even though it was spec'd at 2.5-3-3-6 (IIRC) using the BIOS SPD for memory timings resulted in 3-3-3-8 which are typical for most modern memory these days being used. I manually changed it to the 2.5 numbers and it worked fine.

    Like you, I much prefer reliability over a small increase in speed.
     
Thread Status:
Not open for further replies.
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.