Forum Home
Press F1
Thread ID: 54131	2005-02-04 20:00:00	Expensive Promise FastTrak S150 SX4-M goes poo-poo.	Growly (6)	Press F1

Post ID	Timestamp	Content	User
321614	2005-02-04 20:00:00	Gidday guys, After waiting months for my precious Promise FastTrak to arrive, I encountered several problems with actually using it: </p> It would not boot with my existing Windows XP Pro installation (that was running off onboard SATA RAID). It would not boot Windows Setup to install onto a new RAID array (on the promise card). It would not let me do anything, at all, in my main PC. I tried reserving IRQs, changing PCI slots, swapping cables, swapping drives, asking it nicely, changing boot orders, disabling the onboard SATA controller, resetting CMOS, and even went so far as to bake it a cake, but nothing worked. After calling Taiwan, I was told that I needed to check the thing in another motherboard because it mthe problem may be due to IRQ conflicts. Willing, I put it in my Linux box - in which it booted (although it had conflicts in there, too, but I let that go for now...) In the Linux Box, I recompiled the 2.6.10 kernel (running Debian) to include low-level SATA support (under the SCSI group), specifically the Promise SX4 and TX2 drivers (although not SX4-M, hmmm). Upon booting with a new RAID 0 array, the thing timed out constantly and gave all sorts of whack messages (to follow) when trying to address the drive. So, depressed, I took it out and put in another 200GB SATA 'Cuda (filled with many precious files), but didn't assign it to an Array. It booted without problems the first time, and managed to let me cfdisk it, but it came back and told me that it didn't recognise the filesystem (it was NTFS). So i recompiled the kernel to include NTFS support, and once again it is throwing spazzes. Below is the boot log: WWServer02 kernel: libata version 1.10 loaded. WWServer02 kernel: sata_sx4 version 0.7 WWServer02 kernel: ACPI: PCI interrupt 0000:00:08.0[A] -> GSI 17 (level, low) -> IRQ 17 WWServer02 kernel: Local DIMM ECC Enabled WWServer02 kernel: ata1: SATA max UDMA/133 cmd 0xF8940200 ctl 0xF8940238 bmdma 0x0 irq 17 WWServer02 kernel: ata2: SATA max UDMA/133 cmd 0xF8940280 ctl 0xF89402B8 bmdma 0x0 irq 17 WWServer02 kernel: ata3: SATA max UDMA/133 cmd 0xF8940300 ctl 0xF8940338 bmdma 0x0 irq 17 WWServer02 kernel: ata4: SATA max UDMA/133 cmd 0xF8940380 ctl 0xF89403B8 bmdma 0x0 irq 17 WWServer02 kernel: ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:207f WWServer02 kernel: ata1: dev 0 ATA, max UDMA/133, 390721968 sectors: lba48 WWServer02 kernel: ata1: dev 0 configured for UDMA/133 WWServer02 kernel: scsi0 : sata_sx4 WWServer02 kernel: ATA: abnormal status 0x7F on port 0xF894029C WWServer02 kernel: scsi1 : sata_sx4 WWServer02 kernel: ATA: abnormal status 0x7F on port 0xF894031C WWServer02 kernel: scsi2 : sata_sx4 WWServer02 kernel: ATA: abnormal status 0x7F on port 0xF894039C WWServer02 kernel: scsi3 : sata_sx4 WWServer02 kernel: Vendor: ATA Model: ST3200822AS Rev: 3.01 WWServer02 kernel: Type: Direct-Access ANSI SCSI revision: 05 WWServer02 kernel: SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB) WWServer02 kernel: SCSI device sda: drive cache: write back WWServer02 kernel: SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB) WWServer02 kernel: SCSI device sda: drive cache: write back WWServer02 kernel: sda: unknown partition table WWServer02 kernel: Attached scsi disk sda at scsi0, channel 0, id 0, lun 0 WWServer02 kernel: Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0, type 0 And this is the funny error every time I try and access the thing (located at /dev/ sda: ) WWServer02 kernel: ata1: command timeout WWServer02 kernel: ata1: status=0x51 { DriveReady SeekComplete Error } WWServer02 kernel: SCSI error : <0 0 0 0> return code = 0x8000002 WWServer02 kernel: EOM ILI Current sda: sense = 70 7e WWServer02 kernel: ASC=30 ASCQ=c0 WWServer02 kernel: end_request: I/O error, dev sda, sector 40 WWServer02 kernel: Buffer I/O error on device sda, logical block 5 I have googled the problem, and found similar cases, but they had no clear resolve. I know it's alot to read, and I thank you very much for your time - any help in solving the matter, or pointing me in the right direction would be very very much appreciated. Thanks in advance. /edit: Tidied log outputs I believe this line is of particular significance: WWServer02 kernel: sda: unknown partition table	Growly (6)
321615	2005-02-04 23:14:00	couple of general rules. you can't boot off an existing install. you need to set the raid up, partiton, format then install from new. very few raid controllers will work when moving a disk from another controller. also what type of pci slots does the mobo have as its very easy to flood the pci bus with these cards.	tweak'e (69)
321616	2005-02-05 09:58:00	Apologies, I should've made myself clear: It, as in the computer, would not boot at all with the RAID Controller plugged in - the RAID was still attached to the onboard system. I could not format ANYTHING because the computer froze, before loading anything more than the device listings. It has 32bit PCI slots - and I doubt that's the issue :( At no point did I attempt to boot off an existing RAID with my new controller.	Growly (6)
321617	2005-02-05 10:12:00	sorry to hear . . . . . . there are a pricey item tho raid5 would be very nice :) i take it you would have done the basic's and stripped the system to the basic's . tried different slots? disabled the onboard raid? does it only freeze the OS ? you can get into the bios and raid setup? if your useing on it on 33mhz pci slots you will (with any decent harddrive) flood the pci bus so running other cards may be interesting ;-)	tweak'e (69)
321618	2005-02-05 10:42:00	I've tried: reserving IRQs, changing PCI slots, swapping cables, swapping drives, asking it nicely, changing boot orders, disabling the onboard SATA controller, resetting CMOS, and even went so far as to bake it a cake, but nothing worked. It freezes directly after the device IRQ listings. It POSTs, and I can get as far as entering the card's BIOS - but no futher than Boot from CD: ... Funny that you say I can flood my PCI bus, considering that my existing onboard RAID was also on the PCI bus, and that traditionally IDE controllers are on the PCI bus also... meh.	Growly (6)
321619	2005-02-05 11:15:00	Oh, yeah they are 66MHz :D	Growly (6)
321620	2005-02-05 23:52:00	looks like you've done everthing bar hit it with a big hammer ! what motherboard are you useing?	tweak'e (69)
321621	2005-02-10 15:54:00	Gidday guys, After waiting months for my precious Promise FastTrak to arrive, I encountered several problems with actually using it: . . . . I've exactly the same problems with promixse fasttrak S150 SX4 on hp proliant ml110 and 2 maxtor disks 6Y160M0 . When i try to use disks, kernel says "connection time out, status=0x51, blah blah . . . " as in your PC . Have you found some solution ? thank you	gurutech (7150)
321622	2005-02-16 11:59:00	same as above, exactly like gurutech. After some efforts I finally found something that works: if you install the drives in the inverse order, (like 1st one on slot 3, 2nd on slot 2, etc), the drives are working without errors, (but I don't know why) This is not a solution, but a workaround. I conctacted hp, waiting for a response. HTH	sentenza (7151)
321623	2005-05-18 11:07:00	I now have many further woes with using this card and the onboard controller at the same time - but that's a different matter. Since posting this thread, I've found out that the XOR hardware acceleration chip that the Promise SX4-M has is unsupported by Linux. For this reason, any number of single, standalone disks will work perfectly, but when in RAID they just won't. Bastards.	Growly (6)
1 2