none
Cannot get past disk error on Home Server RRS feed

  • Question

  • When I initiate a backup or when a backup is automatically initiated, I get the following message. :

    Failure Details for Volume (Local Disk):

    An error occurred reading or writing from server storage.
    D:\folders\{00008086-058D-4C89-AB57-A7F909A47AB4}
    /5DOLLARPC-2.C.VolumeClusterLatest2.4096.dat
    ReadFile
    1117

    I have run Checkdsk d: /f /r, but this did not change the results.  Has anyone else had a similar problem that they were able to fix.

    Thanks in advance for your help,

    $5

    Wednesday, September 1, 2010 3:39 PM

Answers

  • Hi,

    Running checkdisk on D: is not enough. Please see this thread on how to check all drives in your server.

    - Theo.


    No home server like Home Server
    • Proposed as answer by kariya21Moderator Thursday, September 2, 2010 4:09 AM
    • Marked as answer by 5Dollar Wednesday, September 8, 2010 1:20 AM
    Wednesday, September 1, 2010 5:23 PM
    Moderator

All replies

  • Hi,

    Running checkdisk on D: is not enough. Please see this thread on how to check all drives in your server.

    - Theo.


    No home server like Home Server
    • Proposed as answer by kariya21Moderator Thursday, September 2, 2010 4:09 AM
    • Marked as answer by 5Dollar Wednesday, September 8, 2010 1:20 AM
    Wednesday, September 1, 2010 5:23 PM
    Moderator
  • When I checked the Chkdsk results, the C: drive had not been checked.  The D: drive did not show any errors.  I am now running the chkdsk on the C: drive.  Will let you know the results.

    5Dollar

    Thursday, September 2, 2010 6:42 PM
  • There is more...
    You are forgetting to check C:\fs\* (see the thread I refered to) ?

    - Theo.


    No home server like Home Server
    Friday, September 3, 2010 8:52 PM
    Moderator
  • Theo,

    Thanks for hanging in there with me. 

    I went to the thread that you suggested and entered the 'sheckall.cmd' as suggested.  For some reason, the C: drive could not be loced, and it stopped there.  I manually entered the command "Chkdsk c: /x /r" and rebooted.  When chkdsk executed, it found no errors on c:.

    I did not see anything about checking c:\fs\*.*.  I just did that and did not find any log or error files. 

    I did find 8 Junctions on 4 different dates, none of which are recent. 

    I have now initiated a new backup, and will let you know how it goes.

    Out of curiosity, what is stored in the fs directory?

    Thanks,

    5Dollar

    Friday, September 3, 2010 9:17 PM
  • Backup failed at the same place 6% with the same error as in first message.  Frustration continues.  Any other ideas?

    5Dollar

    Friday, September 3, 2010 11:37 PM
  • Follow the link Theo provided. Cut and paste the entire .cmd file you will find there, and run it on your server. It will check every drive in your server for errors. As you've seen, you will need to reboot for the C: drive. Once all drives have been checked, reboot your server, which will check C:.

    If any errors are reported, it most likely means you have a failing disk. You should plan to replace it ASAP.

    Whether errors are reported or not, the error you reported in your first post tells me that your backup database is damaged. After replacing a failing disk (if needed) you should use the Repair function in the console to check your database for errors and repair them. Note that this may cause the loss of one or more backups, or possibly all backups for one PC or all backups for all PCs depending on what file is affected. 


    I'm not on the WHS team, I just post a lot. :)
    Saturday, September 4, 2010 2:12 PM
    Moderator
  • I will do this tomorrow as I am out of town.  Here is the cmd file I ran on the server.

    net stop pdl

    net stop whsbackup

    chkdsk D: /x /r

    chkdsk C: /x /r

    for /d %%1 in (C:\fs\*) do start chkdsk /x /r %%1

    I did see many sessions of chkdsk, but did not know what I was looking at.  I thought that multiple tries were being made to check c:.  By Monday, I should be able to run the cmd file.

    Thanks for your help,

    5Dollar

    Sunday, September 5, 2010 12:43 AM
  • That's the file.

    First it stops a couple of critical Windows Home Server services that will throw tons of errors if they're allowed to remain running through the chkdsk process. Next, it runs chkdsk with the /x parameter to dismount the partition prior to checking for errors (which allows the check to proceed immediately) and the /r parameter (surface scan and repair) on D: first, then on C:. C: is the system partition, which is always in use and can't be dismounted, so you're prompted to run on reboot. Answer yes and it will proceed to the for ... command. That runs chkdsk with the same parameters as on D: on the mount points for any additional server storage disks. It runs all of those in parallel; this reduces (somewhat) the elapsed time to check all disks.

    You reboot after for two reasons. First, the check of C: won't occur until the reboot, and you want C: checked. Second, the services that were shut down at the beginning of the script aren't restarted at the end (because you want to reboot immediately for the check of C:).

    After your server is up and running again (assuming it is), the FAQ Theo linked tells you where to find the chkdsk reports in the event logs.


    I'm not on the WHS team, I just post a lot. :)
    Sunday, September 5, 2010 2:37 PM
    Moderator
  • Great explanation!  I see that these other Chkdsk sessions are run in parallel, so that explains the multiple open chkdsk's.  

    What is stored in the 'fs' directory?  What are the 'junctions' that are referenced in this directory?  I assume I get to the logs via the administrative functions.  Can I then send the log files to this forum?  

    After the bad drive is located, can I then try to dismount it and see if that works.  I would then take the drive out of server and replace it with another and 'Add' it to the server.

    Thanks again for your help.

    5Dollar

    Sunday, September 5, 2010 3:06 PM
  • Nothing is stored in the FS folder. The folders under it are links to the other drives in server storage.

    Logs: they're standard event logs, with (in this case) chkdsk reports minus the counters chkdsk uses to let you know it's working. You shouldn't need to paste anything in here; the chkdsk reports are really pretty obvious in my opinion. But yes, you can copy/paste out of one if really needed.

    After you've identified the drive, you can remove it from server storage, yes. That may succeed, if chkdsk reports that it was able to correct all errors, or it may fail, if the hard drive continues to throw frequent errors. If it succeeds, the hard drive was successfully removed, and all files were successfully migrated to other disks. Proceed with repairing the backup database.

    If the removal fails, you will shut down your server, physically remove the drive, restart your server, and remove the "missing" drive from server storage. In this case, make sure that the "missing" drive is the one that was actually causing the problem in the first place before you remove it from server storage. You may see warnings that you'll lose files and/or backups; these aren't uncommon. The former (files) will occur if Windows Home Server has saved any files on the failing drive which are in shares that aren't duplicated. You may be able to recover these files "after the fact" by following the instructions in this FAQ, but if not, they were (to be honest) lost before you ever started down this path. (Choosing not to use duplication for a share is an acceptance that you could potentially lose the entire share if a drive fails in your server.)


    I'm not on the WHS team, I just post a lot. :)
    Sunday, September 5, 2010 3:23 PM
    Moderator
  • I ran the chkdsk cmd file as suggested and did not get any errors.  I did get some information from the event logs that might help solve this problem.

    First, I get a VolSnap 16 error, then multiple, about a 100, viamraid 117 errors.  I also get a Fault Bucket error.  I can post screen captures of these errors but I do not see a way to do it.  How can I send this information to the group?

    5Dollar

     

    Tuesday, September 7, 2010 4:01 PM
  • After running the script in the thread, all appears to be working.  I have now completed 2 backups without any issues.  No disk drives were shown to be bad, and all files appear to be present.

    Thanks for all your help,

    5Dollar

    Wednesday, September 8, 2010 1:23 AM