USB keyboard unresponsive on fresh install

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1313497

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

Hi Brad,
On 28/04/14 14:30, Brad Figg wrote:
> This bug is missing log files that will aid in diagnosing the problem.
>>From a terminal window please run:
>
> apport-collect 1313497
>
> and then change the status of the bug to 'Confirmed'.

I have just done this, you should see the attachments now. Due to being
the reporter of the problem I am unable to mark the bug as "confirmed".

Regards,
--
Stuart Longland
Systems Engineer
     _ ___
\ /|_) | T: +61 7 3535 9619
 \/ | \ | 38b Douglas Street F: +61 7 3535 9699
   SYSTEMS Milton QLD 4064 http://www.vrt.com.au

This change was made by a bot.

Question is maybe whether this is the usb keyboard or maybe some failure to start a local tty or framebuffer mode. If there is a login prompt, framebuffer and tty are most likely ok. If you can ssh into the server from somewhere close to it (to be able to press keys on the keybord at the same time) you could try to install input-utils. "sudo lsinput" should have an entry for the keyboard since it is detected. Then you could "sudo input-events <#>" where number is the number from the /dev line from lsinput.
Look I would have two input lines for my Logitech keyboard here, if that is the case try both. At least one of them should work...

Hi Stefan,
On 29/04/14 01:23, Stefan Bader wrote:
> Question is maybe whether this is the usb keyboard or maybe some
> failure to start a local tty or framebuffer mode. If there is a login
> prompt, framebuffer and tty are most likely ok. If you can ssh into
> the server from somewhere close to it (to be able to press keys on
> the keybord at the same time) you could try to install input-utils.
> "sudo lsinput" should have an entry for the keyboard since it is
> detected.

Well, I had a look, and oddly enough the keyboard is *not* detected.
> root@host:~# lsinput
> /dev/input/event0
> bustype : BUS_HOST
> vendor : 0x0
> product : 0x1
> version : 0
> name : "Power Button"
> phys : "PNP0C0C/button/input0"
> bits ev : EV_SYN EV_KEY
>
> /dev/input/event1
> bustype : BUS_HOST
> vendor : 0x0
> product : 0x1
> version : 0
> name : "Power Button"
> phys : "LNXPWRBN/button/input0"
> bits ev : EV_SYN EV_KEY

There's no entry for the keyboard. I had a poke around but couldn't see
any kernel modules missing. Even after re-plugging the keyboard, it
doesn't appear.

Digging around, I see the USB HID support is a module:
> root@host:~# grep USB_HID /boot/config-3.13.0-24-generic
> CONFIG_USB_HID=m
> CONFIG_USB_HIDDEV=y

Yet, the list of modules on the system seems incomplete:
> root@host:~# find /lib/modules -type f -name \*.ko
> /lib/modules/3.13.0-24-generic/kernel/net/ieee802154/ieee802154.ko
> /lib/modules/3.13.0-24-generic/kernel/net/ieee802154/af_802154.ko
> /lib/modules/3.13.0-24-generic/kernel/net/ieee802154/6lowpan.ko
> /lib/modules/3.13.0-24-generic/kernel/net/netrom/netrom.ko
> /lib/modules/3.13.0-24-generic/kernel/net/x25/x25.ko
> /lib/modules/3.13.0-24-generic/kernel/net/can/can-raw.ko
> /lib/modules/3.13.0-24-generic/kernel/net/can/can.ko
> /lib/modules/3.13.0-24-generic/kernel/net/can/can-bcm.ko
> /lib/modules/3.13.0-24-generic/kernel/net/can/can-gw.ko
> /lib/modules/3.13.0-24-generic/kernel/net/ceph/libceph.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/act_mirred.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_codel.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/em_meta.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_red.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_multiq.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/act_simple.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_dsmark.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_cbq.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/cls_fw.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/cls_bpf.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/cls_cgroup.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/em_nbyte.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_hfsc.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/cls_tcindex.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_choke.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/act_police.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_atm.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_qfq.ko
> /lib/modules/3.13.0-24-generic/kernel/net/sched/sch_plug.ko
> /lib/...

Hi Stuart,

On 29.04.2014 05:23, Stuart Longland wrote:
> Did the linux-3.13.0-24-generic package miss out a few files or are they
> in a separate package the installer forgot to install?

That could be and depends on you seed. For bare-metal you should have seleced
"linux-server" for "d-i base-installer/kernel/image". In general the linux-image
should not be picked directly as we indeed split up the modules into the minimum
for virtual machines and a rest. linux-server, linux-generic, and linux-virtual
are all meta-package which will install the right set of packages and also make
sure you see kernel updates. server and generic nowadays is actually the same.
So you could have installed without those and now miss the linux-image-extra
package.

On 29/04/14 18:35, Stefan Bader wrote:
> Hi Stuart,
>
> On 29.04.2014 05:23, Stuart Longland wrote:
>> Did the linux-3.13.0-24-generic package miss out a few files or are they
>> in a separate package the installer forgot to install?
>
> That could be and depends on you seed. For bare-metal you should have seleced
> "linux-server" for "d-i base-installer/kernel/image". In general the linux-image
> should not be picked directly as we indeed split up the modules into the minimum
> for virtual machines and a rest. linux-server, linux-generic, and linux-virtual
> are all meta-package which will install the right set of packages and also make
> sure you see kernel updates. server and generic nowadays is actually the same.
> So you could have installed without those and now miss the linux-image-extra
> package.

Ahh, I have in my preseed:
> # The kernel image (meta) package to be installed; "none" can be used if no
> # is to be installed.
> d-i base-installer/kernel/image string linux-image-virtual

Perhaps I mis-interpreted the intent of "virtual". Is there a document
that describes the image options in detail that one could refer to?
Perhaps mention of it should be made in the preseed examples?

It's easy to miss the bit about it being for virtual instances in the
dpkg -l output, which is how I installed the image:
> root@host:~# dpkg -l linux-image-virtual
> Desired=Unknown/Install/Remove/Purge/Hold
> | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
> |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
> ||/ Name Version Architecture Description
> +++-=======================-================-================-====================================================
> ii linux-image-virtual 3.13.0.24.29 amd64 This package will always depend on the latest minima

I shall amend my preseed scripts and perhaps we can work on the
documentation (which either is lacking or I missed seeing it).

Regards,
--
Stuart Longland
Systems Engineer
     _ ___
\ /|_) | T: +61 7 3535 9619
 \/ | \ | 38b Douglas Street F: +61 7 3535 9699
   SYSTEMS Milton QLD 4064 http://www.vrt.com.au

I am trying to clean up the bug reports I am assigned to. This one sounds like it got resolved by changing the preseed in order to get the correct set of modules installed. Closing it invalid which may sound a bit harsh but it did not seem to be a bug in the kernel package after all.