TSB12LV23+Oxford 911+IDE woes
Brought to you by:
aeb,
bencollins
Menu
▾
▴
#38 TSB12LV23+Oxford 911+IDE woes
kernel: 2.4.18 (from RH's kernel-source-2.4.18-0.19)
i1394 version: anonymous cvs from this morning
libraw version: 0.9.0
driver: ochi1394
adaptor:http://commerce.www.ibm.com/cgi-bin/ncommerce/ProductDisplay?cntrfnbr=1&prmenbr=1&prnbr=19K5679&cntry=840&lang=en_US
adaptor chipset:
i1394<=>IDE:http://www.zeehoo.com/miva/merchant.mv?Screen=PROD&Store_Code=zeehoo&Product_Code=firewirecase&Category_Code=A
drive:http://www.seagate.com/support/disc/specs/ata/st360021a.html
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
The symptoms are:
(1) cardbus insertion seems to be fine with this output:
----------------------------------------------------
Apr 18 14:34:49 localhost kernel: cs: cb_alloc(bus 3):
vendor 0x104c, device 0x8019
Apr 18 14:34:49 localhost kernel: PCI: Enabling device
03:00.0 (0000 -> 0002)
Apr 18 14:34:49 localhost cardmgr[832]: socket 0:
CardBus hotplug device
Apr 18 14:34:50 localhost /etc/hotplug/pci.agent: Setup
ohci1394 for PCI slot 03:00.0
Apr 18 14:34:50 localhost kernel: ohci1394: $Revision:
1.101 $ Ben Collins <bcollins@debian.org>
Apr 18 14:34:50 localhost kernel: PCI: Setting latency
timer of device 03:00.0 to 64
Apr 18 14:34:50 localhost kernel: ohci1394_0: OHCI-1394
1.0 (PCI): IRQ=[11] MMIO=[c0400000-c0400800] Max
Packet=[2048]
Apr 18 14:34:50 localhost /etc/hotplug/pci.agent:
missing kernel or user mode driver ohci1394
(2) Connecting the firewire disk seems OK, with this
output:
----------------------------------------------------
Apr 18 14:35:48 localhost kernel: ohci1394_0: Error in
reception of SelfID packets [0x80020004/0x0002466e]
(count: 0)
Apr 18 14:35:50 localhost /etc/hotplug/ieee1394.agent:
Setup sbp2 for IEEE1394 product 0x0030e0/0x00609e/0x010483
Apr 18 14:35:50 localhost kernel: ieee1394: sbp2:
Logged into SBP-2 device
Apr 18 14:35:50 localhost kernel: ieee1394: sbp2:
Node[00:1023]: Max speed [S400] - Max payload [2048]
Apr 18 14:35:50 localhost kernel: scsi1 : IEEE-1394
SBP-2 protocol driver (host: ohci1394)
Apr 18 14:35:50 localhost kernel: SBP-2 module load
options:
Apr 18 14:35:50 localhost kernel: - Max speed
supported: S400
Apr 18 14:35:50 localhost kernel: - Max sectors per I/O
supported: 255
Apr 18 14:35:50 localhost kernel: - Max outstanding
commands supported: 8
Apr 18 14:35:50 localhost kernel: - Max outstanding
commands per lun supported:
1
Apr 18 14:35:50 localhost kernel: - Serialized I/O
(debug): no
Apr 18 14:35:50 localhost kernel: Vendor: ST360021
Model: A Rev:
Apr 18 14:35:50 localhost kernel: Type:
Direct-Access ANSI SCSI revision: 06
Apr 18 14:35:50 localhost /etc/hotplug/ieee1394.agent:
missing kernel or user mode driver sbp2
(3) I can use fdisk to partition the drive, this has
always worked and has never caused a problem
----------------------------------------------------
Apr 18 14:36:43 localhost kernel: Attached scsi disk
sda at scsi1, channel 0, id 0, lun 0
Apr 18 14:36:43 localhost kernel: SCSI device sda:
117231408 512-byte hdwr sectors (60022 MB)
Apr 18 14:36:43 localhost kernel: sda: sda1 < sda5
sda6 sda7 sda8 sda9 sda10 sda11 sda12 >
(4) mke2fs seems by all indications to be doing
something, writing superblocks, and generally spinning
and accessing the disk
(5) whenever I try to mount the partition (formatted
with mke2fs) I get the following error:
----------------------------------------------------
Apr 18 14:37:47 localhost kernel: EXT2-fs error (device
sd(8,5)): ext2_check_descriptors: Block bitmap for
group 1 not in group (block 0)!
Apr 18 14:37:47 localhost kernel: EXT2-fs: group
descriptors corrupted!
(6) if I format the partitions with the IDE drive
connected directly to the IDE bus of a desktop machine,
I can then mount & use the partition. I CANNOT,
however, unmount and then remount (never disconnecting
neither the firewire plug nor the cardbus card, doing
nothing with the scsi bus, writing nothing to the
partition, nothing). I can mount them only once.
----------------------------------------------------
(7) once, over the course of a week of messing with
this and wasting hours it happened that I could
mount/unmount repeatedly without errors. This
occurrence has not repeated itself
----------------------------------------------------
Can you help? Do you have any suggestions? I'm
considering buying different hardware and trying it.
Would you recommend getting a second i1394 cardbus card
or a second disk+i1394 combo?
Thanks very much,
Darrell Schiebel
Logged In: YES
user_id=196071
I'm half tempted to believe this is a bug of sorts in the
hardware. Maybe James will have more information (assigning
this to him).
Logged In: YES
user_id=519318
I verified that this drive (seagate ide + oxford911) worked
fine on a windows 2000 system. It appears, can be formatted
& used under win2k. Also, back under linux, the vfat
formatted partition does not exhibit these problems, i.e. it
can be written to, unmounted, remounted, etc.
Logged In: YES
user_id=519318
I just noticed that in fdisk you can see the tweaking of the
partition data that happens as a result of
mounting/unmounting these partitions:
------------------------------------------------------------
Disk /dev/sda: 255 heads, 63 sectors, 7297 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 1275 10241406 c Win95
FAT32 (LBA)
/dev/sda2 1276 2550 10241437+ 83 Linux
/dev/sda3 2551 3825 10241437+ 83 Linux
/dev/sda4 3826 7297 27888840 83 Linux
------------------------------------------------------------
The first three partitions started at the same size, but
after mounting/unmounting the block count changes.
Logged In: YES
user_id=519318
I'm not sure about this last comment (2002-04-19 14:20).
While it is true that the three partitions started as the
same size, perhaps Win2K which changed the size (of one)
rather than linux (changing the size of two).
Logged In: YES
user_id=519318
I find that this cardbus ieee1394 adaptor & oxford 911
bridge & ide drive work fine on a (much) older Dell Inspiron
3200! I installed RH 7.3b2 (2.4.18) on both machines. Now
the Inspiron can mount & use the drive with no problems, but
the IBM X22 is still out of luck. Both seem to have a
"yenta_socket" cardbus bridge. On the IBM X22 there are:
02:03.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev 80)
02:03.1 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev 80)
on the old Inspiron 3200 there are:
00:04:0 CardBus bridge: Texas Instruments PCI1131 (rev 01)
00:04:1 CardBus bridge: Texas Instruments PCI1131 (rev 01)
Can anyone offer any suggestions? Is this likely defective
hardware or hardware which is functioning properly but which
has "features" which the driver code fails to account for? I
assume the later since I have a modem (3com 3cxm756) pccard
which works fine.
thanks, Darrell
PS: when I eject the ieee1394 card from either laptop, the
laptop reboots...
Logged In: NO
I have had the same reboot problem with Mandrake 8.2
(2.4.18) when even putting same CD back into drive.
After a lot of trial and error, I finally was able to see
and mount a 1394 CDROM (via Oxford 911 bridge) and browse
with Konqueror. BUT when I even just open and then shut
door the system would reboot.
My hardware is Abit KG7 and Actiontec (LUCENT) IEEE1394 PCI
Card.
I am just trying to see if 1394 will even work.. and I've
tried a lot of things.. so my configuation is unknown..
Modules ieee1394, ohci1394, and raw1394 are now installed
but I did not rebuild kernal. Still after reboot I must
reenter modprobe ohci1394 and modporbe raw1394 so I can see
the drive and access the files (Note, the mount point and
hardware configuration is again picked up and mounts shown).
Logged In: YES
user_id=519318
I tried a second disk (EZQuest), same behavior. So to recap:
o PCMCIA card & both disks work in a DELL 3200
o PCMCIA card & both disks fail to work in IBM X22
o have used other PCMCIA cards (MHz 3CXM756)
This stinks! ... thanks for listening...