April 13, 201016 yr Hello, I am having an unusual and disturbing problem when trying to add a couple of new disks to my unraid system. I have one of the original MD-1200 that has been upgraded with new components - SUPERMICRO MBD-C2SEE-O LGA 775 Intel G43 ATX Intel Motherboard, a couple of the super micro 5 in 3 SATA enclosures, and a SUPERMICRO AOC-SAT2-MV8 64-bit PCI-X133MHz SATA Controller Card. I kept the 2 promise IDE controllers from the original system, and currently have 6x500G IDE drives hooked up to them, 5x1TB SATA drives, and 2x1.5TB SATA drives in the enclosures. As the motherboard only has 6 SATA ports, one of the drives has been running off the SATA card. I am running Unraid v4.5.3 and using a couple of add ons: Unmenu v1.1 and preclear_disk.sh v.9.8. I have had great success with version .9 of preclear_disk, and I don't believe at all that upgrading that is causing my issues... I just picked up a couple of Samsung 2TB drives (which happen to be on sale with promo code through tomorrow - http://www.newegg.com/Product/Product.aspx?Item=N82E16822152202) When I start the preclear, I get errors like the following in my syslog: Apr 13 18:54:51 BigMama kernel: hdl: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:51 BigMama kernel: hdl: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:51 BigMama kernel: hdl: possibly failed opcode: 0xa1 Apr 13 18:54:51 BigMama kernel: hdj: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:51 BigMama kernel: hdj: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:51 BigMama kernel: hdj: possibly failed opcode: 0xa1 Apr 13 18:54:51 BigMama kernel: hde: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:51 BigMama kernel: hde: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:51 BigMama kernel: hde: possibly failed opcode: 0xa1 Apr 13 18:54:52 BigMama kernel: hdg: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:52 BigMama kernel: hdg: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:52 BigMama kernel: hdg: possibly failed opcode: 0xa1 Apr 13 18:54:52 BigMama kernel: hdh: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:52 BigMama kernel: hdh: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:52 BigMama kernel: hdh: possibly failed opcode: 0xa1 Apr 13 18:54:52 BigMama kernel: hdf: task_pio_intr: status=0x51 { DriveReady SeekComplete Error } Apr 13 18:54:52 BigMama kernel: hdf: task_pio_intr: error=0x51 { UncorrectableError SectorIdNotFound AddrMarkNotFound }, LBAsect=86864678551812, sector=18446744073709551615 Apr 13 18:54:52 BigMama kernel: hdf: possibly failed opcode: 0xa1 If I try to play a video off of one of these drives it won't start, and doing an ls -l on a directory gives me garbage (bunch of ? etc). If I kill the preclear, shut down the array, and reboot everything is fine. Video plays as always, dir looks great, and I don't get errors like this until I start pre-clear again. I have 2 power supplies with plenty of juice, so that isn't the problem. I reseated cards and cables... I'm at a real loss here since only the pre-clear causes this issue to surface. I see both drives and have tried preclearing both with the same results. I find it hard to believe that both drives arrived from Newegg with problems that would lead to this... I've placed a copy of my syslog here - http://pastebin.com/pV5p45DN and the output of dmesg here - http://pastebin.com/VsuL33vQ Any thoughts? I'm at a loss... Thanks much! Jeff Edited: I should add that I've run hdparm on the new drives, spun up/down other drives, basically everything I can think of to push things. I just finished a parity check yesterday to prepare for this, and everything came out ok then. Very strange.
April 13, 201016 yr You have the disk controller in an emulated IDE mode? Why? The two disks should be showing up as /dev/sdX, not dev/hdX. I do not think you want to use the emulated IDE mode. Set them to AHCI mode. Other than that, the errors seem to indicate it was emulating a pio mode? In any case, one drive can affect the other on an IDE emulation. Joe L.
April 13, 201016 yr Author You have the disk controller in an emulated IDE mode? Why? The two disks should be showing up as /dev/sdX, not dev/hdX. I do not think you want to use the emulated IDE mode. Set them to AHCI mode. Other than that, the errors seem to indicate it was emulating a pio mode? In any case, one drive can affect the other on an IDE emulation. Joe L. The 2 new drives show up as /dev/sda and /dev/sdc. That snippet from the syslog is the older IDE drives which are the ones that report the issue... preclear keeps running, but I'm a bit scared to let it go since the other drives report these problems.
April 16, 201016 yr Author So I powered everything down and physically removed all the IDE drives, once booted without the array running I ran the pre-clear script on both drives. It finished on both new drives with nothing significant to report, took > 24 hrs per drive but that part is over When I started to upgrade the parity disk, I got hit with the unformatted bug I've read about. So I cancelled the parity rebuild, stopped array, restarted, and now parity is rebuilding. The thought of another large chunk of time without the array was more than I could take! I just hope it finishes without incident... I've had good luck over the years.
Archived
This topic is now archived and is closed to further replies.