usb-mass storage disk isn't recognized the right way

i have to correct one thing: despite the dmesg output looks good for the above chassis, the sda1/2 whatever device nodes aren't created.

but when mounted into another chassis i posess, the disk works without problems.

lsusb
Bus 004 Device 024: ID 04b4:6830 Cypress Semiconductor Corp. USB-2.0 IDE Adapter

problem still persists with new dmesg output:

[4303799.894000] usb 4-4: new high speed USB device using ehci_hcd and address 5[4303800.337000] Initializing USB Mass Storage driver...
[4303800.338000] scsi0 : SCSI emulation for USB Mass Storage devices
[4303800.339000] usb-storage: device found at 5
[4303800.339000] usb-storage: waiting for device to settle before scanning
[4303800.339000] usbcore: registered new driver usb-storage
[4303800.340000] USB Mass Storage support registered.
[4303805.340000] Vendor: ST312002 Model: 6A Rev: 3.06
[4303805.340000] Type: Direct-Access ANSI SCSI revision: 00
[4303805.347000] usb-storage: device scan complete
[4303806.043000] Driver 'sd' needs updating - please use bus_type methods
[4303806.045000] SCSI device sda: 234441649 512-byte hdwr sectors (120034 MB)
[4303806.046000] sda: assuming drive cache: write through
[4303806.049000] SCSI device sda: 234441649 512-byte hdwr sectors (120034 MB)
[4303806.049000] sda: assuming drive cache: write through
[4303806.049000] sda: sda1 sda2 sda3
[4303806.067000] sd 0:0:0:0: Attached scsi disk sda
[4303806.090000] sd 0:0:0:0: Attached scsi generic sg0 type 0

checked again today with fresh installs of ubuntu 5.10 and 6.06 final. Harddisk works with 5.10. 6.06 doesn't work, same dmesg as above. Harddisk Access-LEDS glows permanently.

The PL3507 is widely known for a weak FireWire implementation and sometimes extremely bad firmware. USB is supposed to work better on this though.

As far as sbp2 is concerned, failures to connect are perhaps fixed upstream. There was a regression in sbp2 concerning some PL3507 in Linux 2.6.16...2.6.16.19 inclusive.

The regression I referred to in comment 4 went into Ubuntu Dapper too, as part of a bugfix for other devices. Ubuntu Edgy has the regression fixed:
http://www.kernel.org/git/?p=linux/kernel/git/bcollins/ubuntu-2.6.git;a=commitdiff;h=a54c9d30dbb06391ec4422aaf0e1dc2c8c53bd3e

Fixes related to this went into dapper-security. Please test it.

Thanks for releasing a fix for this serious problem. Unfortunately, the fix does not seem to work for me.

I made a standard system upgrade. After that, I have 'uname -srv' "Linux 2.6.15-27-k7 #1 SMP PREEMPT Sat Sep 16 02:35:20 UTC 2006", 'grep DESC /etc/lsb-release' 'DISTRIB_DESCRIPTION="Ubuntu 6.06.1 LTS"'.

No lines are printed on /var/log/dmesg or /var/log/syslog when plugging and unplugging usb. 'sudo rmmod ehci_hcd' does not help. Devices are not created either: 'ls /dev/sd*' "ls: /dev/sd*: No such file or directory". USB key lights are not lit when it is connected. The same USB key works on another Dapper machine with older set of updates.

My USB mouse works normally. 'lsusb' shows exactly the same printout with and without USB key plugged in.
$ lsusb # id hidden by me "[..]"
Bus 001 Device 004: ID [..] Logitech, Inc.
Bus 001 Device 001: ID 0000:0000
Bus 002 Device 001: ID 0000:0000

This bug refers to two problems. One in firewire and the other in usb.

Can you test that it does or doesn't work with firewire?

I don't have test machines with Firewire, so unfortunately I can't test it. Can someone else try this with Firewire?

I tested that it is not working with USB mass storage. I think USB problem is more serious, as it is much more common.