Looks like I got myself into trouble! My NFS disk server failed to come up after being shutdown (properly) at night
(Power supply dead). So I took the ATA133 Maxtor disk and its UDMA PCI controller,
Promise 20269 chipset and moved it to another system, Redhat 8, 2.4.28 kernel.
There is only one Ext3 partition on the drive, full 120GB, and it came up
as hde1. I was able to mount it locally and work with it for a while. Then
I noticed an extrordinary amount of disk action, and when that subsided the
mount directory was empty. I could not unmount the drive partition and it
said the drive was busy and unavailable, so I shutdown the system at the
console, and restarted the computer again.

The Promise controller couldn't even start the drive and started scrolling
sector errors, and clicking sounds came from the drive. So I took the drive
to another system with a built on motherboard Promise 20267 raid controller
( I am not using RAID) Same deal, but no clicking sounds.
So I used hdparm to disable drive dma, and set it to read only.
I am able to access the drive with fdisk and see the partition table, but
I get scrolling sector errors when I try to mount the drive partition. Ideas
anyone? I have to rescue this drive. Can I do a "dd" to another drive?

I tried fsck on the partition. Here is the /var/log/messages tail, if it helps

May 10 13:05:05 infoserve kernel: hdf: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
May 10 13:05:05 infoserve kernel: hdf: read_intr: error=0x01 { AddrMarkNotFound }, LBAsect=63, sector=0
May 10 13:05:06 infoserve kernel: hdf: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
May 10 13:05:06 infoserve kernel: hdf: read_intr: error=0x40 { UncorrectableError }, LBAsect=63, sector=0
May 10 13:05:06 infoserve kernel: end_request: I/O error, dev 21:41 (hdf), sector 0
May 10 13:05:08 infoserve kernel: hdf: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
May 10 13:05:08 infoserve kernel: hdf: read_intr: error=0x40 { UncorrectableError }, LBAsect=64, sector=1
May 10 13:05:08 infoserve kernel: end_request: I/O error, dev 21:41 (hdf), sector 1
May 10 13:05:09 infoserve kernel: hdf: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
May 10 13:05:09 infoserve kernel: hdf: read_intr: error=0x40 { UncorrectableError }, LBAsect=65, sector=2
May 10 13:05:09 infoserve kernel: end_request: I/O error, dev 21:41 (hdf), sector 2
May 10 13:05:11 infoserve kernel: hdf: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
May 10 13:05:11 infoserve kernel: hdf: read_intr: error=0x40 { UncorrectableError }, LBAsect=66, sector=3
May 10 13:05:11 infoserve kernel: end_request: I/O error, dev 21:41 (hdf), sector 3

Thanks for any suggestions.
Sri

Yea, I'm getting that too. I'm wondering if there is a set of hdparm parameters or kernel idebus= arguments that could make this go away. Any ideas anyone?

Another thread on this is here: http://www.linuxquestions.org/questi...ght=dma_expiry

Another suggestion that I read about was that it may be a power problem, in that the power supply couldn't keep up with that many hard disks.

Well, I've configured the start up scripts such that the drives connected to the extra 3 Promise controllers are running udma3 rather than the udma5 or 6 that the drives are capable of. I figure that this slower DMA speed would allow the controller to keep up with the timing, in case there is DMA contention on the PCI bus. Let you know if it works, and someone posts a reply.

Changing the UDMA setting for the drives didn't work either.

Instead, I took the third Promise Controller card out of the system, and made use of the secondary IDE controller from the motherboard.

Since I removed the third controller, I've not had any errors, DMA or otherwise. Seems like you can only put two Promise IDE controllers in a single system. Guess I'll have to start looking as some sort of distributed file system to scale the storage that I'd like to get to.