Firewire and USB disks both want SCSI /dev/sda

Crito · 01-02-2006, 04:15 PM

I don't know if it's hardware or software or distro specific, but figured I'd start here as hardware was all I was playing with when the trouble started.

I have external firewire hard disks that were working fine with FC4 before I bought an internal compact flash card reader that works off USB. It seems both the fireware and USB "mass storage" drivers use SCSI emulation and want to be /dev/sda leaving me with a system where I can either use one or the other but not both. It's really annoying me at this point. I thought SCSI emulation went the way of the dinosaur in the 2.6 kernel, but guess that must have just been for CD/DVD burners.

Anyway, right now automount is broken for firewire and if I mount a firewire drive manually at CLI it locks up my USB card reader until I do a cold boot (a warm reboot won't fix it.) Any advice on how to get firewire and USB mass storage devices coexisting peacfully would be greatly appreciated.

ilikejam · 01-03-2006, 01:49 PM

Hi.

All USB and Firewire drives use SCSI emulation (well, it's not really emulation, more SCSI over USB/Firewire).

Each drive should take the next free sdX letter when plugged in, so there really shouldn't be a problem. Plug them in one at a time, and post the last, say, 30 lines of 'dmesg' when they're both plugged in so we can see what's going on.

Dave

Crito · 01-03-2006, 07:06 PM

Wouldn't you know it, seems to be working now. I *think* what the problem is that the internal flash drive has four individual readers. In Windows all four devices are mapped to drive letters immediately, even without any media inserted. In Linux it appears /sda /sdb /sdc /sdd are only assigned when media is in the drive. In any case, my firewire drive shows up as /sde if I mount it after a flash card. If I mount the firewire drive first it shows up as /sda and the internal reader locks up. Anyway here are some relevant lines from dmesg|tail that might help. TIA

Code:

SCSI device sda: 250368 512-byte hdwr sectors (128 MB)
sda: Write Protect is off
sda: Mode Sense: 02 00 00 00
sda: assuming drive cache: write through
SCSI device sda: 250368 512-byte hdwr sectors (128 MB)
sda: Write Protect is off
sda: Mode Sense: 02 00 00 00
sda: assuming drive cache: write through
 sda: sda1
SELinux: initialized (dev sda1, type vfat), uses genfs_contexts
ieee1394: Current remote IRM is not 1394a-2000 compliant, resetting...
ieee1394: Node added: ID:BUS[0-00:1023]  GUID[0000000e0000175b]
ieee1394: Node changed: 0-00:1023 -> 0-01:1023
sbp2: $Rev: 1306 $ Ben Collins <bcollins@debian.org>
ieee1394: sbp2: Driver forced to serialize I/O (serialize_io=1)
ieee1394: sbp2: Try serialize_io=0 for better performance
scsi1 : SCSI emulation for IEEE-1394 SBP-2 Devices
ieee1394: sbp2: Logged into SBP-2 device
ieee1394: Node 0-00:1023: Max speed [S400] - Max payload [2048]
  Vendor: HDS72258  Model: 0VLAT20           Rev:
  Type:   Direct-Access-RBC                  ANSI SCSI revision: 04
SCSI device sde: 160836480 512-byte hdwr sectors (82348 MB)
sde: asking for cache data failed
sde: assuming drive cache: write through
SCSI device sde: 160836480 512-byte hdwr sectors (82348 MB)
sde: asking for cache data failed
sde: assuming drive cache: write through
 sde: sde1
Attached scsi disk sde at scsi1, channel 0, id 0, lun 0
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 0
Attached scsi generic sg1 at scsi0, channel 0, id 0, lun 1,  type 0
Attached scsi generic sg2 at scsi0, channel 0, id 0, lun 2,  type 0
Attached scsi generic sg3 at scsi0, channel 0, id 0, lun 3,  type 0
Attached scsi generic sg4 at scsi1, channel 0, id 0, lun 0,  type 14
Device not ready. Make sure there is a disc in the drive.
Device not ready. Make sure there is a disc in the drive.
Device not ready. Make sure there is a disc in the drive.
Device not ready. Make sure there is a disc in the drive.
Device not ready. Make sure there is a disc in the drive.
Device not ready. Make sure there is a disc in the drive.
program python is using a deprecated SCSI ioctl, please convert it to SG_IO
sda: Write Protect is off
sda: Mode Sense: 02 00 00 00
sda: assuming drive cache: write through
SCSI device sda: 250368 512-byte hdwr sectors (128 MB)
sda: Write Protect is off
sda: Mode Sense: 02 00 00 00
sda: assuming drive cache: write through
 sda: sda1
SELinux: initialized (dev sda1, type vfat), uses genfs_contexts

ilikejam · 01-04-2006, 02:53 PM

Does the reader lock up if you mount from the command line?

Crito · 01-05-2006, 09:32 PM

The USB drives automount on my default FC4 install. The firewire drives used to automount too, but since installing the internal reader I've had to mount manually. Only other thing I see involved in the equation is:

Code:

[surfer@localhost ~]$ ps aux|grep usb-storage
root      1527  0.0  0.0      0     0 ?        S    Jan03   0:32 [usb-storage]
surfer    7314  0.0  0.1   3768   680 pts/1    S+   22:24   0:00 grep usb-storage

But like I said before, as long as I remember to insert a disk in the internal USB reader's drive before mounting a firewire drive everything seems to work (though I'm mounting firewire manually now.) So my immediate problem is solved, thanks. Would be nice if someone made them play together better though, but it's useable (if not very user friendly.)

EDIT: just to be complete, here's the relevant info about my USB controller and internal reader from command
cat /proc/bus/usb/devices

Code:

T:  Bus=01 Lev=00 Prnt=00 Port=00 Cnt=00 Dev#=  1 Spd=480 MxCh= 8
B:  Alloc=  0/800 us ( 0%), #Int=  0, #Iso=  0
D:  Ver= 2.00 Cls=09(hub  ) Sub=00 Prot=01 MxPS= 8 #Cfgs=  1
P:  Vendor=0000 ProdID=0000 Rev= 2.06
S:  Manufacturer=Linux 2.6.14-1.1653_FC4 ehci_hcd
S:  Product=EHCI Host Controller
S:  SerialNumber=0000:00:10.4
C:* #Ifs= 1 Cfg#= 1 Atr=e0 MxPwr=  0mA
I:  If#= 0 Alt= 0 #EPs= 1 Cls=09(hub  ) Sub=00 Prot=00 Driver=hub
E:  Ad=81(I) Atr=03(Int.) MxPS=   2 Ivl=256ms

T:  Bus=01 Lev=01 Prnt=01 Port=04 Cnt=01 Dev#=  3 Spd=480 MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=11b0 ProdID=6787 Rev= 0.05
S:  Manufacturer=Atech Flash.
S:  Product=PRO-IX
S:  SerialNumber=427051747166
C:* #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=500mA
I:  If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
E:  Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms

ilikejam · 01-06-2006, 03:54 AM

You might want to open a bug report on the Fedora Bugzilla. There really shouldn't be any conflicts with this sort of thing.

Dave

Crito · 04-25-2006, 11:04 AM

Believe it or not I finally resolved this issue. It's distro specific apparently. Had I installed my AFT PRO - 9 flash reader under Mandriva, for example, it would have worked fine. Fedora Core, however, does not install or re-run Anaconda after initial installation. I've actually replaced an entire motherboard (with different brand chipset even) under Mandrake 9.2 and upon next reboot everything all my hardware was re-detected and re-installed automagically. Apparently kudzu, which they both have in common, just does configuration chages. Anyway, long story short, I just installed FC5 on same machine and now have all my devices straightened out.

P.S. Is it me or has google search engine become craptacular lately. Example, I'm trying to search for this thread there (because LQ's database has poor indexes and searching here takes forever -- but that's a different pet peeve..) and entered the following keywords: "firewire linux crito". Google gives me edu sites with crito in the domain name, even if content of site itself doesn't contain some of the keywords at all!! So I enter the exact same keywords in http://www.altavista.com/ and bingo, there's what I'm looking for.

If google keeps futzing up their search engine I'm going to start telling people to "altavista it".

Crito · 04-25-2006, 11:21 AM

Just remembered in Mandrake/Mandriva the equivalent of Anaconda is called HardDrake. So to summarize: on Fedora Core Anaconda is not installed or run after intial setup and on Mandrake HardDrake is installed and re-run on every boot.

I suppose that makes Mandrake more flexible, on the one hand, but makes Fedora Core boot a lot faster, on the other.