January 29, 201115 yr I have 3 2tb EARS drives I purchased recently. I needed to format them in my desktop before preclearing them on my server due to some weird shenanigans with it reporting only 1tb. Those issues have been fixed, but the NEW issue seems to be that when I formatted them in my desktop, my gigabyte motherboard (WHILE NOT NOTING ANYTHING IN THE BIOS) installed an HPA on all three. Now with the latest bios installed it has an option to disable HPA, but the damage is done and all three now have HPA on them. 2 of the drives are currently in the array, 1 is a data drive and 1 is parity. the third is connected to the system but hasn't gone through preclear yet.Anything I can do? some images and screenshots attached, along with syslog. Right now the drive in the data drive slow is unimportant so if I need to hose the data on the parity and data drive that is fine. DASH: http://dl.dropbox.com/u/519591/UNRAID/dash.JPG SYSLOG: http://dl.dropbox.com/u/519591/UNRAID/syslog-2011-01-29.txt Errors marked by Unmenu Jan 29 09:23:21 Tower kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 300) (Drive related) Jan 29 09:23:21 Tower kernel: ata4.00: HPA detected: current 3907027055, native 3907029168 (Errors) Jan 29 09:23:21 Tower kernel: ata4.00: ATA-8: WDC WD20EARS-00MVWB0, 51.0AB51, max UDMA/133 (Drive related) Jan 29 09:23:21 Tower kernel: ata4.00: 3907027055 sectors, multi 0: LBA48 NCQ (depth 31/32) (Drive related) Jan 29 09:23:21 Tower kernel: ata4.00: configured for UDMA/133 (Drive related) Jan 29 09:23:21 Tower kernel: ata5.00: HPA detected: current 3907027055, native 3907029168 (Errors) Jan 29 09:23:21 Tower kernel: ata5.00: ATA-8: WDC WD20EARS-00MVWB0, 51.0AB51, max UDMA/133 (Drive related) Jan 29 09:23:21 Tower kernel: ata5.00: 3907027055 sectors, multi 0: LBA48 NCQ (depth 0/32) (Drive related) Jan 29 09:23:21 Tower kernel: ata5.01: HPA detected: current 3907027055, native 3907029168 (Errors) Jan 29 09:23:21 Tower kernel: ata5.01: ATA-8: WDC WD20EARS-00MVWB0, 51.0AB51, max UDMA/133 (Drive related) Jan 29 09:23:21 Tower kernel: ata5.01: 3907027055 sectors, multi 0: LBA48 NCQ (depth 0/32) (Drive related) Jan 29 09:23:21 Tower kernel: ata5.00: configured for UDMA/133 (Drive related) Jan 29 09:23:21 Tower kernel: ata5.01: configured for UDMA/133 (Drive related) I'm not sure if this is a related issue, but I have noticed my system likes to hang temporarily now when transferring files to the array. Here is a screenshot of my bandwidth when transferring a 6GB file to disk 5, you can see how it drops twice. http://dl.dropbox.com/u/519591/UNRAID/BW.JPG Thanks everyone for the help.
January 29, 201115 yr Author shameless self reply. Looks like my next steps are to use hdparm -N p3907029168 /dev/XYZ on all three drives since they are identical discs? My question now is how do I get back to normal operation? Do I toast the data drive and tell it to rebuild, and then toast the parity drive and rebuild parity drive again? Or since I dont care about the contents of the data disk, can I use HDPARM on both, and then just hit the button to start building a new parity? What about the one not in the array, can I just do HDPARM and then do preclear like normal? I still have that issue with my transfer drops too, but hopefully this would fix it? thanks all!
January 29, 201115 yr shameless self reply. Looks like my next steps are to use hdparm -N p3907029168 /dev/XYZ on all three drives since they are identical discs? My question now is how do I get back to normal operation? Do I toast the data drive and tell it to rebuild, and then toast the parity drive and rebuild parity drive again? Or since I dont care about the contents of the data disk, can I use HDPARM on both, and then just hit the button to start building a new parity? What about the one not in the array, can I just do HDPARM and then do preclear like normal? I still have that issue with my transfer drops too, but hopefully this would fix it? thanks all! You need to increase the size of the parity disk first, otherwise the array will not start. Whatever you do, only do ONE drive at a time, otherwise you will not be able to start the array. Is your server motherboard a Gigabyte Motherboard?? Or just the one in the desktop PC? Joe L.
January 29, 201115 yr Author Server Motherboard is fine, its the desktop that was causing problems. my 1TB drives that have never touched the desktop are totally fine. Sigh. Ok I will go 1 at a time per your reccomendation. Thanks Joe! Any thoughts on my weird networking problems? I dont see issues in the syslog when transferring, so I am at a loss. EDIT: Looks like something went wrong. I used the command and rebooted, and now the drive shows up as 1.5TB. Should I use an external tool instead? http://dl.dropbox.com/u/519591/UNRAID/post%20HDPARM%20result.JPG
January 29, 201115 yr Server Motherboard is fine, its the desktop that was causing problems. my 1TB drives that have never touched the desktop are totally fine. Sigh. Ok I will go 1 at a time per your reccomendation. Thanks Joe! Any thoughts on my weird networking problems? I dont see issues in the syslog when transferring, so I am at a loss. EDIT: Looks like something went wrong. I used the command and rebooted, and now the drive shows up as 1.5TB. Should I use an external tool instead? http://dl.dropbox.com/u/519591/UNRAID/post%20HDPARM%20result.JPG My first guess is you made a typo when typing the hdparm command. You only get to set the size once per power cycle, you might need to power down and back up to try again. Joe L.
January 29, 201115 yr Author Tried twice copy/paste, (hdparm -N p3907029168 /dev/sde) and ended up getting the same result each time. Think I should try seatools?
January 29, 201115 yr Tried twice copy/paste, (hdparm -N p3907029168 /dev/sde) and ended up getting the same result each time. Think I should try seatools? give it a try.
January 29, 201115 yr Author Looks like it worked. *552 is that sweet spot on new drives right? http://dl.dropbox.com/u/519591/UNRAID/FIXED.JPG I did it to the parity drive and the drive outside of the array. Next steps are finish this party rebuild, then do the same to the data disk looks like. Man, what a mess. After this I need to find out why my connectivity is dropping!
Archived
This topic is now archived and is closed to further replies.