locked
Backup stalls at 35% RRS feed

  • Question

  •  

    I'm trying to back up my Vista Home Premium workstation to a new WHS and the backup stalls every night at about 35%. At first I thought it was due to Vista's reknowned network slowness, but I applied SP1 which boosted transfer rates from 2-3 MB / sec to 7 MB / sec on a 100Mbit wired network. I've also read that backups continue past the end of the window.

     

    I'm trying to back up nearly 200 GB.

     

    Thanks to the background search, I see that Paul Ruskin is having the same problem.

     

    http://forums.microsoft.com/WindowsHomeServer/ShowPost.aspx?PostID=2892023&SiteID=50

     

    The Vista SP1 box is a Shuttle SS31T, Pentium D 3.2 GHz, 2 GB ram, Vista Home Premium, Geforce 8600 GTS, 2x 250 GB SATA drives in a RAID 1 Mirror (onboard SiS fakeRAID controller), and has an addin 4-port USB card.

     

    The WHS is a Pentium D 3.2 GHz, Gigabyte motherboard, 2GB RAM, 4x 250 GB drives in a RAID 5 array controlled by a 3Ware controller.

     

    Both systems are fully patched via Microsoft Update.
    Friday, February 29, 2008 10:01 AM

All replies

  • Have you run chkdsk on the drives?  That has fixed some backup issues in the past.
    Friday, February 29, 2008 10:04 AM
  • I had this problem as well.

     

    I ran chkdsk on the local drive, and watched the output.

     

    It showed two files with chkdsk errors.  I deleted each file off the hard drive, and then chkdsked again to verify there were no more errors.

     

    Restored the files from the WHS backups (the good ones).  Ran chkdsk to make sure the errors didn't resurface.

     

    Problem fixed.

     

    5 minutes total fix time - Boooyah!

     

    Friday, February 29, 2008 5:06 PM
  • Chkdsk ran with no errors. This morning the backup had again stalled at 35%. I cancelled it and started a manual backup and watched it while doing other work (yay for having two monitors). It proceeded quite happily, with the disk light showing continuous activity until - you guessed it - 35%. Then it simply died.

     

    Does anything special happen at 35%?

     

    I forgot to add: I moved over 50 GB of files from the workstation to the server manually, so I'm doubting that this is a problem with a particular file.

      

     

    Saturday, March 1, 2008 10:41 AM
  • There is, of course, nothing of any use whatsoever in the Event Logs. I can see when the backups start and I can see when I cancel them, but nothing of any actual use.

    Saturday, March 1, 2008 11:10 AM
  • I had the same problem. When I configured the backup in my console I had excluded 3 folders. One was one I created in the root of c called "No Backup". The other 2 were in my documents, My Music and My Videos. Once I moved the 2 into the No Backup folder, it worked fine. I would not promise it will resolve your issue, but visiting any exclusions may help.

     

    Saturday, March 1, 2008 12:13 PM
  •  Quentin Stephens wrote:

     

    The WHS is a Pentium D 3.2 GHz, Gigabyte motherboard, 2GB RAM, 4x 250 GB drives in a RAID 5 array controlled by a 3Ware controller.

     

    Both systems are fully patched via Microsoft Update.

     

    Sorry if I've got this wrong, but It is my understanding that WHS servers are not designed for RAID arrays, 5 or otherwise.

     

    A pain I know, but what about reconfiguring without the RAID array?  In any case 250GB is very small for the primary WHS disk, but should work as long as your backup files aren't too large.

    Saturday, March 1, 2008 12:23 PM
  • Here's another tidbit: I cancelled the backup locally, but the WHS console says the backup is still ongoing and won't cancel it.

     

    Saturday, March 1, 2008 12:31 PM
  • I've just watched the backup from start to stall.

     

    Up to 30%, it was checking which clusters had changed.

     

    The next 5% was 'Reorganising data on the server'

     

    At 35% it became ' Sending data to the server'. There was a short pause followed by a short burst of HDD activity, and since then it's just sat there.

     

    I ran CHKDSK on both ends before starting, and deleted all the faulty earlier backups

    Saturday, March 1, 2008 2:19 PM
  •  Rach3 wrote:
     Quentin Stephens wrote:

     

    The WHS is a Pentium D 3.2 GHz, Gigabyte motherboard, 2GB RAM, 4x 250 GB drives in a RAID 5 array controlled by a 3Ware controller.

     

    Both systems are fully patched via Microsoft Update.

     

    Sorry if I've got this wrong, but It is my understanding that WHS servers are not designed for RAID arrays, 5 or otherwise.

     

    It's a hardware RAID and just appears to Windows as one drive. Since I needed more than 250 GB in one partition I thought it best to use it.

    Saturday, March 1, 2008 2:21 PM
  • I just had a thought..  I upgraded my old WHS server with a gigabit lan card and one of my pc's stopped backing up also...  It was some netgear card..  I switched it out with an intel pro and everything was fine..  Do you have another lan card you could possibly try on your whs?
    Saturday, March 1, 2008 2:38 PM
  •  Enchanter wrote:
    I just had a thought..  I upgraded my old WHS server with a gigabit lan card and one of my pc's stopped backing up also...  It was some netgear card..  I switched it out with an intel pro and everything was fine..  Do you have another lan card you could possibly try on your whs?

     

    I've got plenty of NICs lying around - 3com and Broadcom. Trouble is, the workstation has no free slots. Anyway, I don't know if it's relevant, but I can't actually cancel the backup without rebooting the local machine. If I cancel the backup, the local machine recognises that the backup has been cancelled but the server does not.

    Saturday, March 1, 2008 3:05 PM
  • OK, first of all, you could try deleting the computer from the "Computers and Backups" tab, then, using the WHS Toolkit, reset the entire backup database (effectively wiping it clean).

     

    Then on your Vista box, run c:\program files\windows home server\discovery.exe to re-setup the connector and backup settings, then finally try a manual backup and see if it still gets stuck.

    Saturday, March 1, 2008 3:55 PM
  •  

    I'll add that to the list. In the meantime, I've dug out my games box, a Windows XP SP2 machine and tried to back that up. It's stalled at 39%. So it looks like a problem with either the connector software or the server.

     

    The only way I've managed to get the server to abort a backup is to reboot the local machine.

    Saturday, March 1, 2008 5:40 PM
  •  

    This is dashed wierd: I cancelled the backup on the XP box, and rebooted it to get the server to cancel the session, but blow me down if the backup didn't restart, then proceed up to 44% before the cancellation kicked in.

     

    I ran the Connector Advisor in admin mode and it passed all tests.

    Saturday, March 1, 2008 5:53 PM
  • By the sound of it, with the XP box also failing at the point it starts "sending data to the server", your backup folder on the server is corrupted.

     

    Install the WHS toolkit, and use it to delete the backup database in its entirety, and then run discovery.exe to re-setup the connector software and backups.

     

    Saturday, March 1, 2008 5:55 PM
  •  

    I did that, and it's still failing at 35%. However, using the toolkit, I did turn on verbose logging and look ye here:

     

    [01/03/2008 18:50:33  b48] ReconnectableStream: Read bytes 843055104 up to 843071467, generation 14
    [01/03/2008 18:50:33  b48] ReconnectableStream: Read bytes 843071467 up to 843071488, generation 14

    [01/03/2008 18:50:33  b48] Error decrypting message (8009030f)
    [01/03/2008 18:50:33  b48] Breaking TLS sesssion...
    [01/03/2008 18:50:33  b48] Read 0 bytes, hr=8009030f
    [01/03/2008 18:50:33  b48] ReconnectableStream: Read failed, reconnecting {C355BFD8-3B33-4FCF-9BDD-A5725F6C8DD8}, generation 14
    [01/03/2008 19:03:37  d38] Session: DoGetConfiguration
    [01/03/2008 19:03:37  d38] Session: DoGetConfiguration
    [01/03/2008 19:03:47  d38] Session: DoGetConfiguration
    [01/03/2008 19:03:47  d38] Session: DoGetConfiguration

     

    and it continues to log Session: DoGetConfiguration entries

     

    What does that mean?

    Saturday, March 1, 2008 7:15 PM
  • That error code - 8009030f matches one given in a thread started by Pentahex in July

     

    http://forums.microsoft.com/WindowsHomeServer/ShowPost.aspx?PostID=1766448&SiteID=50

     

    Apparently Chris Gray of the WHS team was able to fix it. I've uploaded the server logs as CAB 448656619.

    Saturday, March 1, 2008 10:41 PM
  • I had this happen to me today on a laptop i turned off system restore on the laptop and it worked.

     

     

    Sunday, March 2, 2008 12:01 AM
  •  tedhersh3 wrote:

    I had this happen to me today on a laptop i turned off system restore on the laptop and it worked.

     

     

    I wonder... I'll try to exclude the System Volme Information directory.

    Sunday, March 2, 2008 8:33 AM
  • Another thought: is it possible that the local client needs to run as a local administrator and isn't? How can I check?

    Sunday, March 2, 2008 9:00 AM
  • I tried excluding a number of the '(Unknown)' directories in WHS - the hidden System Volume Information directory wasn't there and it made no difference. Incidentally, you only seem to be able to exclude top-level directories.

    Sunday, March 2, 2008 9:38 AM
  • I ran the server storage diagnostics and it came up with something in the backup area and I'm not sure what it means:

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Commit.dat

    Attributes: A

    CT :2008\03\01 21:08:36:140

    LAT:2008\03\02 11:41:19:390

    LWT:2008\03\02 11:41:19:390

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Commit.dat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Commit.dat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Commit.dat

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Index.4096.dat

    Attributes: A

    CT :2008\03\01 22:08:24:343

    LAT:2008\03\02 11:39:50:250

    LWT:2008\03\02 11:39:50:250

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Index.4096.dat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Index.4096.dat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\Index.4096.dat

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.BackupSet.dat

    Attributes: A

    CT :2008\03\01 21:14:30:250

    LAT:2008\03\02 11:41:19:406

    LWT:2008\03\02 11:41:19:296

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.BackupSet.dat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.BackupSet.dat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.BackupSet.dat

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.C.Volume.dat

    Attributes: A

    CT :2008\03\01 21:14:30:218

    LAT:2008\03\02 11:41:00:171

    LWT:2008\03\02 11:40:15:062

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.C.Volume.dat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.C.Volume.dat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.C.Volume.dat

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.VolumeConfig.configdat

    Attributes: A

    CT :2008\03\01 21:10:36:656

    LAT:2008\03\02 11:41:00:156

    LWT:2008\03\02 11:40:25:156

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.VolumeConfig.configdat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.VolumeConfig.configdat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\QTSWS4.VolumeConfig.configdat

     

    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\{6F890C44-E00B-484F-BA33-0651FE93184A}.Machine.configdat

    Attributes: A

    CT :2008\03\01 21:10:36:687

    LAT:2008\03\02 11:41:19:406

    LWT:2008\03\02 11:40:25:281

    File can not be queried for reparse point : D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\{6F890C44-E00B-484F-BA33-0651FE93184A}.Machine.configdat

    DeviceIOControl failed D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\{6F890C44-E00B-484F-BA33-0651FE93184A}.Machine.configdat ,Error: 232

    Fail to read tombstone content for file D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}\{6F890C44-E00B-484F-BA33-0651FE93184A}.Machine.configdat

     

    Checking files not in D:\, D:\DE, D:\shares, D:\folders......

    Files in D:\ but not in shares or folders:

    File D:\QSM_VolumeID, size: 16 bytes

    File D:\TCP PArameters.reg, size: 21456 bytes

    Files in D:\shares but not in shares or folders:

    Files in D:\folders but not in shares or folders:

    Files in D:\DE but not in shares or folders:

     

    0 long filenames are found.

     

    Drive D:\, 87980797952 bytes used, 640499347456 bytes left

    Total Space used by storage: 728480145408 bytes

    Total Space left for storage: 640499347456 bytes

    Spaced used by backup: 11757436 bytes

    Spaced used by shares: 174141494866 bytes

    Total Files: 37010

    Files in folders: 7

    Files in shares: 37003

    Files encrypted: 0

    Files cannot open: 0

    Files opened by other programs: 0

    Missing shadow due to missing disk: 0

    Files with both shadows missing: 7

    Shadow ACL diff from Tombstone: 0

    Tombstone ACL doesn't match QSM: 0

    DEMigrator service restore state to 4 pending...

    DEMigrator service restore state to 4 pending...

    DEMigrator service is restored.

    Sunday, March 2, 2008 11:49 AM
  • Quentin, those errors make me question the health of your system drive. There's definitely something seriously wrong with your backup database. Before you go down the "failing drive" path, I would suggest you see if working your way through brubber's database consistency how-to would help. If it doesn't, you can try resetting your backup database.
    Sunday, March 2, 2008 3:30 PM
    Moderator
  • I've run CHKDSK /R /F on both server partitions - and I ran it on the workstation overnight - and deleted all backups. I'm now doing a manual backup.

    Sunday, March 2, 2008 5:05 PM
  •  Quentin Stephens wrote:

    I'm now doing a manual backup.

     

    And it's failed in exactly the same place in exactly the same way: at 35% and with a short burst of activity and then nothing.

     

    I've spent several days on this issue. Hopefully some Microsoft person will browse this forum when they get to work on Monday. I'm leaving it until then.

    Sunday, March 2, 2008 6:14 PM
  •  

    I got an automated email asking if it's been fixed yet. Now, while I'm still having the same problem, I know MS employs good people but I'm not unhappy if they don't work over the weekend!
    Monday, March 3, 2008 4:53 PM
  • I've reinstalled WHS twice today, once a simple reinstall, the second a reinstall with a 3com NIC, and the backup still fails. Both times I deleted the backup data first. I'm now going to try a full reinstall with the 3Com NIC.

     

     

    Tuesday, March 4, 2008 6:26 PM
  • Quentin:

     

    As long as you're doing a full reinstall, why don't you try changeing your RAID array to JBOD.  This operating system was not intended to run with RAID arrays, and will easily handle terabytes of data across multiple drives as if they were one big drive.  The fact that your problems seem related to file system errors makes me think the RAID array could be confusing the operating system.

     

    Regards,

    Merrill

     

    Tuesday, March 4, 2008 6:59 PM
  •  

    The documentation indicates that WHS doesn't like the fakeraid that's part of many PC BIOSes. This is proper hardware RAID. It's just as if I have a single 750 GB HDD. However, if this fails, I will do that. Also, everything works fine apart from the backup.
    Tuesday, March 4, 2008 7:05 PM
  • Well, it failed again.

    Tuesday, March 4, 2008 11:04 PM
  • I rebuilt the server without the 3Ware RAID controller. Two SATA drives Still using the 3com NIC. Still fails at about 35%. So it's not the 3Ware controller.

     

    Now, there are two clients. One Vista, one XP. The XP box is a P4-2400.

     

    I'm wondering: both workstations have fakeraid software RAID - the Vista box uses the onboard SiS controller and the XP box uses an Adaptec 1210. Could that be the problem? Paging Microsoft!

     

     

     

     

    Wednesday, March 5, 2008 1:14 PM
  • Well, I pulled the mirror from my XP box and plugged in a plain IDE drive, installed XP and the connector software, and it backed up just fine.

     

    Wednesday, March 5, 2008 4:40 PM
  • Looks like the next step is to rebuild my workstation from scratch. Deep joy.

    Wednesday, March 5, 2008 5:12 PM
  •  

    I copied my data directory to the server overnight (110 GB). Then I excluded that directory and tried the backup. It got to 35%, but only after exercising the disk for a very long time.
    Thursday, March 6, 2008 1:45 PM
  • Not an answer but a workaround: complete reinstall of each client workstation. This is not good at all. Imagine the technician going to install a WHS at someone's home. He's pre-built the box and instead of having to just plonk the box on the client's network and install the connector on each client PC, he's got to budget (both in time and money) for reinstalling the OS and applications on every single client PC. Not forgetting to not lose any data. That's potentially several days work. This cost will, of course, be passed on to the customer. What was a £100 / $200 surcharge becomes a £500+ / $1000+ surcharge. Bad MS. 

    Monday, March 10, 2008 3:54 PM
  • I will check that out but the drive on the server was just reformated to load the server, and my clinet PC (vista 64) is less tham a month old on both the OS install and disk drive
    Draklor
    Thursday, April 2, 2009 9:48 PM
  • I used create a system image, after many attempts at create a full backup. Worked fine.
    Sunday, November 22, 2009 11:52 PM
  • Thread necrophillia, but I found the NETworx freeware bandwidth monitoring program caused this error.  Remove it and all is well.
    Saturday, April 30, 2011 6:14 AM
  • Thread necrophilia or not, you just found the solution to my problem, my friend!

    Thank you very much for your post, Ivan.

    Once I stopped Networx (I didn't even need to uninstall it), the backup worked again smoothly and no longer stopped at 35%.

    This has had me stumped since March! Thanks again!

    Saturday, May 7, 2011 5:26 PM
  • Another one bitten by Networx.    I don't know why it suddenly stopped backups after not causing a problem for about 3 months.    After 3 days of missed backups on the one client with Networx I found this thread and removed it.   Problem solved...many thanks !!!
    Friday, May 20, 2011 6:25 PM
  • WOW. Good thing I read to the end of this thread. Networx was my problem as well. It was hard to diagnose as one laptop would work sometimes and not others. Also everything worked fine for at least a month or two before I started having problems.

    Thanks for figuring this one out.

    Saturday, July 30, 2011 1:10 PM