My WHS is sick RRS feed

  • Question

  • All,
    I posted this question on the HTPC section on the AVS Forum and got no response. Maybe someone can give me some pointers here.


    My WHS is home-built on a BioStar Mobo with WHS PP 3 running from a 500 GB WD IDE drive. It is on a GbE LAN. Data are stored in three WD GP drives. 2 x 1TB and 1 x 2 TB. It has been running as trouble-free as a toaster now for 18 months. 24/7/365. File transfer speeds to and from it vary between 300 and 800 Mbs. Backups of the three clients takes about 10-15 minutes each (typicallY) in the middle of the night.
    I use it as a music server, a backup server, a photo server, and overflow storage for "Recorded TV" for our two 4-tuner W7 MC PCs. I have been delighted with its performance until now.
    Last week the 2 TB GP drive "went missing". The "Connector" reported that the shared folders pooled across the 3 drives was corrupted and files were lost. The backup database was also reported as being corrupted. S***!
    A server reboot did not correct. Power and data cables seemed fine. A power down/power up reboot returned the drive to "active duty"! Everything was fine. But I could no longer watch a recorded TV program stored on the server. It stops and starts. File transfers from the "Shared Folders" to and from occur at about 2-5 MegaBytes per second! And on occasion they would report "Trouble accessing drive" during a drag and drop transfer. But not always. Sometimes they would succeed. And backups were taking hours.
    So I "Removed" the 2TB drive from the drive extender data pool. This took 1/2 a day. I luckily had enough free space on the 1 TB drives to consolidate the data. I also did a "repair" on the backup database and it said there were no problems with it.
    No help. Still slow as molasses. I CAN drag a file to and from the WHS Desktop (using RDP) from and to a client shared folder at 800 Mbs. So the network is not the problem. This is of course on the IDE OS drive.
    I also put the 2 TB drive in a client PC and beat it up pretty good. Chkdsk, WD diagnostics, file transfers, etc. No problems whatsoever. S.M.A.R.T. data looks fine as best as I can tell. It's a bit cryptic. Been running in the client for a week now.
    So I picked up a WD 2TB USB drive today and put it on the server to allow me to "Remove" the two 1TB drives and test them. My guess is one of them has gone sour. The "removal" of the first drive should take several hours at the rate the progress indicator is moving.
    My plan is to remove the other one when it is done and see if file transfers to the OS drive and USB drive still seem reasonable.
    I have a 1TB USB drive on this client and it typically transfers at about 350 Mbs. I know it will be about 10x slower than the SATA drives but it should work while I test the drives. I also have some left over 320s and 500s I can put in to make sure the SATA controller and cables are OK.
    Any other suggestions appreciated.


    P.S. The connector has been "Removing" the first 1 TB drive for 14 hours now and it is about 1/3 done. One CPU core is running at 100% on it. Does this make sense?

    Saturday, January 15, 2011 1:21 PM