external USB harddrive produces lots of kernel messages

[Jonas Meurer <jonas@freesources.org>]

hello,

i get USB kernel messages printed to the console regularely:

usb 1-5.4: reset high speed USB device using ehci_hcd and address 3

in past, i used the onboard USB controller with a via chipset, and
additional to the messages, i had read/write errors. 
then someone told me, that the via USB 2.0 chipset is known to be
bad quality. So i bought an Adaptec USB 2.0 controller with a NEC
chipset.

now, lspci gives the following:
0000:00:08.0 USB Controller: NEC Corporation USB (rev 43)
0000:00:08.1 USB Controller: NEC Corporation USB (rev 43)
0000:00:08.2 USB Controller: NEC Corporation USB 2.0 (rev 04)

i've connected two USB 2.0 devices to the controller:
- an external USB 2.0 400GB harddisk (Seagate)
- a USB 2.0 multi- cardreader

the kernel messages are produced by the external harddisk.

fortunately, the read/write errors disappeared with the new NEC USB
controller, but the kernel messages still occur.

i still get the following message inregularely, but the average may
be something like 3 minutes. often the message is printed twice at once:

usb 1-5.4: reset high speed USB device using ehci_hcd and address 3

and another log message is printed to the console quite inregularely:

ehci_hcd 0000:00:08.2: qh ffff81001f00aa00 (#00) state 4


unfortunately these messages mess up fullscreen console apps like mutt,
irssi, vim etc. but that's not the biggest problem. i could configure
syslog-ng to just print them to kern.log and not to the current console.

but what i really want is to know why these messages occur, and how i
fix that.

do you think the the external harddrive is the problem, regardless which
controller i use? if yes, is there a way to fix it?

i'dd really love to see these messages disappear.

greetings
 jonas