REGRESSION: USB mouse & keyboard don't work immediatly on login

[Geoffrey De Smet]

My Microsoft non-wireless mouse & keyboard worked perfectly on jaunty (9.04), until I installed all the latest updates last friday (2009-06-19). Those new updates of a little over 2 weeks old, included a kernell update it seems, because ubuntu had to restart.

Since then, 2 major regressions are introduced:
- the startup takes a lot longer. At some point the ubuntu logo dissappears and I get console messages about USB. It prints one, has a certain time out, then prints the next and so on, 8 times in general (4 for each device I think)
- both the USB mouse and keyboard don't work immediately in the login screen. After the login screen is shown for about a minute, they all of a sudden start working.
Unplugging the mouse and keyboard and plugin them in when the login screen is visible, is a (although annoying) workaround.

During shutdown the ubuntu logo dissappears again and I get these messages:
usb 5-1: device description read/64 error -110
usb 5-1: device description read/64 error -110
usb 5-1: device description read/64 error -110
usb 5-1: device description read/64 error -110
usb 5-1: device description read/8 error -110
usb 5-1: device description read/8 error -110
usb 5-1: device description read/8 error -110
usb 5-1: device description read/8 error -110

I had exactly the same problem on my old pc, with the same Microsoft USB mouse on ubuntu 8.04. I figured Ubuntu 8.10 and later finally fixed it once and for all.
Even if there is some magical command to fix my problem today, the update shouldn't cause this problem and it should be ensured (through unit tests or something) that it doesn't happen again on the next update.
Normal users are not able to find and run such magical commands, nor are they tolerant to such regressions upon installing updates.

What files do I need to attach to this issue so you can decently evaluate this issue?

[Geoffrey De Smet (ge0ffrey)]

xorg.conf hasn't been changed since 2009-04-25, so the regression update of 2009-06-19 didn't change it.

[Geoffrey De Smet (ge0ffrey)]

From kern.log, take a look at the times which clearly prove that the startup procedure now takes minutes instead of seconds:

...
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 77.138453] EXT3 FS on sda6, internal journal
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.249211] type=1505 audit(1245576986.563:2): operation="profile_load" name="/usr/share/gdm/guest-session/Xsession" name2="default" pid=2110
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.285868] type=1505 audit(1245576986.599:3): operation="profile_load" name="/sbin/dhclient-script" name2="default" pid=2114
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.285952] type=1505 audit(1245576986.599:4): operation="profile_load" name="/sbin/dhclient3" name2="default" pid=2114
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.285986] type=1505 audit(1245576986.599:5): operation="profile_load" name="/usr/lib/NetworkManager/nm-dhcp-client.action" name2="default" pid=2114
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.286016] type=1505 audit(1245576986.599:6): operation="profile_load" name="/usr/lib/connman/scripts/dhclient-script" name2="default" pid=2114
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.397123] type=1505 audit(1245576986.708:7): operation="profile_load" name="/usr/lib/cups/backend/cups-pdf" name2="default" pid=2119
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.397269] type=1505 audit(1245576986.708:8): operation="profile_load" name="/usr/sbin/cupsd" name2="default" pid=2119
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.457370] type=1505 audit(1245576986.771:9): operation="profile_load" name="/usr/sbin/mysqld" name2="default" pid=2123
Jun 21 11:36:27 ge0ffrey2008 kernel: [ 78.478204] type=1505 audit(1245576986.792:10): operation="profile_load" name="/usr/sbin/tcpdump" name2="default" pid=2127
Jun 21 11:36:32 ge0ffrey2008 kernel: [ 83.950838] usb 1-5: device descriptor read/64, error -110
Jun 21 11:36:34 ge0ffrey2008 kernel: [ 85.119695] ppdev: user-space parallel port driver
Jun 21 11:36:37 ge0ffrey2008 kernel: [ 88.700179] e1000e 0000:00:19.0: irq 2301 for MSI/MSI-X
Jun 21 11:36:37 ge0ffrey2008 kernel: [ 88.756049] e1000e 0000:00:19.0: irq 2301 for MSI/MSI-X
Jun 21 11:36:37 ge0ffrey2008 kernel: [ 88.756296] ADDRCONF(NETDEV_UP): eth0: link is not ready
Jun 21 11:36:38 ge0ffrey2008 kernel: [ 90.284764] 0000:00:19.0: eth0: Link is Up 100 Mbps Full Duplex, Flow Control: None
Jun 21 11:36:38 ge0ffrey2008 kernel: [ 90.284767] 0000:00:19.0: eth0: 10/100 speed: disabling TSO
Jun 21 11:36:38 ge0ffrey2008 kernel: [ 90.284934] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Jun 21 11:36:47 ge0ffrey2008 kernel: [ 99.164015] usb 1-5: device descriptor read/64, error -110
Jun 21 11:36:47 ge0ffrey2008 kernel: [ 99.380513] usb 1-5: new high speed USB device using ehci_hcd and address 4
Jun 21 11:36:48 ge0ffrey2008 kernel: [ 100.576006] eth0: no IPv6 routers present
Jun 21 11:37:02 ge0ffrey2008 kernel: [ 114.492011] usb 1-5: device descriptor read/64, error -110
Jun 21 11:37:17 ge0ffrey2008 kernel: [ 129.708011] usb 1-5: device descriptor read/64, error -110
Jun 21 11:37:17 ge0ffrey2008 kernel: [ 129.924009] usb 1-5: new high speed USB device using ehci_hcd and address 5
Jun 21 11:37:22 ge0ffrey20...

[Geoffrey De Smet (ge0ffrey)]

The assigned package could be wrong.

[Geoffrey De Smet (ge0ffrey)]

This issue fixed itself with the new linux header that updated yesterday evening. Thanks :)