Seagate harddrive errors

Message

riven · Post by **riven** » Thu Apr 28, 2005 4:45 pm

Hey all. I got a new Seagate SATA 2000.7 200GB harddrive the other day. Its on an Sli 3112 controller at the moment (so performance sucks) the errata fix comes on ok, but I'm still getting some errors.

ata1: status=0x51 { DriveReady SeekComplete Error }
ata1: error=0x04 { DriveStatusError }

These appear in dmesg just over time (it might be when the drive is accessed not sure). Do they look dangerous? Anyway to rid me of them?

I shall be getting a new motherboard soon (hence having the cheapest crapest sata card money could buy just for now) will be NF4 so i will use that for the sata. Just worried about data integrity in the mean time.

Cheers anyone.

GBob · Post by **GBob** » Thu Apr 28, 2005 4:50 pm

Two things, if your power supply is old, get a new one. (You may hear a slight beep from the hard drive when these messages appear, I did.) Second, 5 year warranty. If a new power supply doesn't fix things, send it back.

riven · Post by **riven** » Thu Apr 28, 2005 4:56 pm

Powersupply is fine, 485W enermax thats about a month old and is only powering a 2400XP+, 1Gig ram Radeon 9000, 2xHD 2xOptical harddrives. Do you recon its a faulty harddrive then?

GBob · Post by **GBob** » Thu Apr 28, 2005 7:03 pm

install the smartd tools and look at the error rate. If the drive is reporting errors that it is failing, it would make a good case for sending it back.

riven · Post by **riven** » Thu Apr 28, 2005 7:21 pm

Device: /dev/sda, opened
Device /dev/sda, SATA disks accessed via libata are not currently supported by
smartmontools. When libata is given an ATA pass-thru ioctl() then an
additional '-d libata' device type will be added to smartmontools.

Doesnt look like thats going to work

I think i'll email seagate

matador · Post by **matador** » Fri May 13, 2005 11:56 am

Seagate has a good tool on their homepage (similar to sartmontools). Download the Seatools cd and burn it. It results in a 99% reliable result (they say). Also check that you have set CONFIG_IDEDISK_MULTI_MODE in the kernel and other IDE controllers.

riven · Post by **riven** » Fri May 13, 2005 4:32 pm

I contacted Seagate support and they got me to run the tool. It checked out ok, so i guess its just the controller. I will be upgrading my motherboard in a couple of months, so it will do until then.

matador · Post by **matador** » Sat May 14, 2005 2:46 pm

I doubt its your motherboard. You could always try and see what knoppix says, but I think you're missing something in the kernel or that you have a wrong setting in the BIOS (or maybe even hdparm). My humble suggestions would be to double check your kernel (/usr/src/linux/.config) with "lspci", also check "hdparm -I /dev/sda".

I've got a similar problem due to a dma conflict in the BIOS, hence my suspitions. On the other hand, different motherboard = different modules.

riven · Post by **riven** » Sat May 14, 2005 2:56 pm

i think you misunderstood what I meant by upgrading my motherboard. At the moment my motherboard does not have onboard SATA, its only PATA IDE. So to run the SATA drive i am using a very cheap pci SATA controller which uses the Sil3112 chipset. This chipset is know to have problems with seagate drives. So i suspect its the SATA controller card. When I upgrade to a new motherboard, it will have the nvidia sata controller, so should be compatable.

hdparam doesnt work with sata drives (using the SCSI emulation thingymajig).

How do I check for a dma conflict?

vespaman · Post by **vespaman** » Sun May 15, 2005 7:10 am

How often do you get these errors?

I have the same errors in my log, on 4 disks on 2 different sil3114 controllers. My disks are Hitachi 7K250, motherboard a Asus A8N-SLI Deluxe. I have had many problems with this set-up. The rest of the disks are on another SIL3114, a SIL3112, and on the Nforce4, and they seem to run along without any issues. I suspect the disks and cables ...

I got maybe 10 errors last night, copying 1,5tb from one raid to another.

riven · Post by **riven** » Sun May 15, 2005 7:58 am

I get them quite often. I think more often if the disk is doing multiple operations, such as reading and writing. I've also had the another type of error that occasionally occurs during very heavy writing, that causes the disk to pause for about a minute. As far as I can tell I'm not suffering any data corruption, and the Seagate tools say my drive is running fine, but it doesnt recognise my controller.

matador · Post by **matador** » Sun May 15, 2005 7:27 pm

Exams does weired things with people... I don't know SATA so my answers are limited on that front.

Anyway, for the dma check with "lspci" if the first numbers/letters are the same. For instance 0000:00:1f.0, 0000:00:1f.1, 0000:00:1f.2 ...