Random usb failure at bootup

classic Classic list List threaded Threaded
18 messages Options
Reply | Threaded
Open this post in threaded view
|

Random usb failure at bootup

solitone
Hi,

this morning I experienced a bad issue that worried me for a while, but
happily it ended well with no consequences.

After resuming from suspend, the monitor was black, and I had to press the
power button to hard shutdown the computer.

On reboot, the monitor still didn't switch on. I tried several times. In the
last tries the monitor switched on, but the keybord still didn't work. This
affected the boot loader as well. The system booted up, and I got the login
screen, however no key worked (apart from the power button, it's a MacBookPro
12,1).

I brutally switched off the laptop, and leaved off for some time (1 hour).
When I tried and rebooted it, everything worked well again.

I inspected the kernel messages, and found the following usb errors (keyboard
and monitor are USB devices on the MacBookPro 12,1):

====
Mar 29 09:06:25 alan kernel: usb 2-3: device not accepting address 2, error
-62
Mar 29 09:06:36 alan kernel: usb 2-3: device not accepting address 3, error
-62
Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: Stopped the command ring
failed, maybe the host is dead
Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: Abort command ring failed
Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up
Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up
Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up
Mar 29 09:06:56 alan kernel: usb 2-3: device not accepting address 4, error
-108
Mar 29 09:06:56 alan kernel: usb usb2-port3: couldn't allocate usb_device
====

It seems that some rest is sometimes needed ;-)

Reply | Threaded
Open this post in threaded view
|

System broken after yesterday's upgrade

solitone
Hi, something really weird is going on today on my system--debian stretch on
an Apple MacBookPro 12,1.

Some time ago I noticed a strange issue with USB, which prevented the keyboard
and the trackpad to work after bootup (they are both USB devices on
MacBookPro's). Here's an extract of the kernel logs that I collected at the
time:

> Mar 29 09:06:25 alan kernel: usb 2-3: device not accepting address 2, error
> -62
> Mar 29 09:06:36 alan kernel: usb 2-3: device not accepting address 3, error
> -62
> Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: Stopped the command ring
> failed, maybe the host is dead
> Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: Abort command ring
> failed Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died;
> cleaning up Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died;
> cleaning up Mar 29 09:06:55 alan kernel: xhci_hcd 0000:00:14.0: HC died;
> cleaning up Mar 29 09:06:56 alan kernel: usb 2-3: device not accepting
> address 4, error -108
> Mar 29 09:06:56 alan kernel: usb usb2-port3: couldn't allocate usb_device

After some rest, everything started working again, and I didn't worried--till
this morning, when I rebooted! Again, keyboard and trackpad no longer worked.
Again I brutally switched of the laptop, and tried to wait some time, but
nothing, this time it didn't help! I suspect the misbehaviour depends on the
upgrade I carried on yesterday, since this was my first reboot since then.

The really strange thing is that even grub now is sooo slow to show up. It
takes 30 secs or more! Obviously, the keyboard does not work there. After some
additional time, the kernel boots up, but it stalls for 30 seconds on the step
"loading initial ramdisk". After that, it boots up correctly, displays the
login manager, but as I said I have no keyboard nor trackpad.

I tried many times. In the end I tried and plugged external usb mouse and
keyboard. After bootup they didn't work either, but after some additional 30
seconds they started working, together with the integrated keyboard and the
trackpad. No idea what's going on.

I rebooted without the external usb devices plugged in, and waited for several
minutes, but nothing happened, keyboard and trackpad didn't start working. I
rebooted again, like before with the two external usb devices attached, and
like before everything started working after some pretty long time.

As I said, the culprit might be yesteday's upgrade. Among others, the kernel
was upgraded:
 
Start-Date: 2017-04-11  07:33:56
Commandline: packagekit role='update-packages'
Requested-By: solitone (1000)
Upgrade: [...] linux-image-4.9.0-2-amd64:amd64 (4.9.13-1, 4.9.18-1
) [...]

Everything worked fine before. Now the first steps (starting from grub) in the
bootup process are unbelievably slow, and the laptop's keyboard and trackpad
don't work unless I wait much time and do some weird ritual, like plugging in
some external usb devices and again wait and wait.

Now everything works well, but I fear that at the next reboot I'll have to
struggle, hope, and pray that I can use my sytem again!

Here are the kernel error message that I've collected during the last reboot:

solitone@alan:~$ sudo journalctl -k -b -1 -p 3
-- Logs begin at Sun 2017-02-12 21:33:16 CET, end at Wed 2017-04-12 15:42:11
CEST. --
Apr 12 14:42:14 alan kernel: brcmfmac 0000:03:00.0: firmware: failed to load
brcm/brcmfmac43602-pcie.txt (-2)
Apr 12 14:42:14 alan kernel: brcmfmac: brcmf_c_preinit_dcmds: Firmware version
= wl0: Nov 10 2015 06:38:10 version 7.35.177.61 (r598657) FWID 01-ea662a8c
Apr 12 14:42:14 alan kernel: brcmfmac: brcmf_cfg80211_reg_notifier: not a
ISO3166 code (0x30 0x30)
Apr 12 14:42:15 alan kernel: brcmfmac: brcmf_p2p_create_p2pdev: set p2p_disc
error
Apr 12 14:42:15 alan kernel: brcmfmac: brcmf_cfg80211_add_iface: add iface
p2p-dev-wlp3s0 type 10 failed: err=-16
Apr 12 14:42:21 alan kernel: brcmfmac: brcmf_inetaddr_changed: fail to get arp
ip table err:-23
Apr 12 14:42:22 alan kernel: usb 2-3: device not accepting address 2, error
-62
Apr 12 14:42:33 alan kernel: usb 2-3: device not accepting address 3, error
-62
Apr 12 14:42:39 alan kernel: Bluetooth: hci0 urb ffff8ae9df53b840 failed to
resubmit (22)
Apr 12 14:42:39 alan kernel: Bluetooth: hci0 urb ffff8ae9df53b000 failed to
resubmit (22)
Apr 12 14:42:39 alan kernel: Bluetooth: hci0 urb ffff8ae9df53b900 failed to
resubmit (22)
Apr 12 14:42:39 alan kernel: Bluetooth: hci0 urb ffff8ae9df53bf00 failed to
resubmit (22)
Apr 12 14:42:39 alan kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up
Apr 12 14:42:39 alan kernel: bcm5974 1-5:1.2: could not read from device
Apr 12 14:46:12 alan kernel: INFO: task kworker/1:1:46 blocked for more than
120 seconds.
Apr 12 14:46:12 alan kernel:       Not tainted 4.9.0-2-amd64 #1
Apr 12 14:46:12 alan kernel: "echo 0 > /proc/sys/kernel/
hung_task_timeout_secs" disables this message.
Apr 12 14:48:12 alan kernel: INFO: task kworker/1:1:46 blocked for more than
120 seconds.
Apr 12 14:48:12 alan kernel:       Not tainted 4.9.0-2-amd64 #1
Apr 12 14:48:12 alan kernel: "echo 0 > /proc/sys/kernel/
hung_task_timeout_secs" disables this message.

As expected, kworker complaints depends on XHCI:

solitone@alan:~$ sudo journalctl -k -b -1 -p 4
[...]
Apr 12 14:48:12 alan kernel: INFO: task kworker/1:1:46 blocked for more than
120 seconds.
Apr 12 14:48:12 alan kernel:       Not tainted 4.9.0-2-amd64 #1
Apr 12 14:48:12 alan kernel: "echo 0 > /proc/sys/kernel/
hung_task_timeout_secs" disables this message.
Apr 12 14:48:12 alan kernel:  ffff8ae9e2d43c00 0000000000000000
ffff8ae9e4857040 ffff8ae9eec981c0
Apr 12 14:48:12 alan kernel:  ffff8ae9e4da0000 ffffa26c00dffb70
ffffffff973f6403 000000005480022c
Apr 12 14:48:12 alan kernel:  00ff8ae9e4da0000 ffff8ae9eec981c0
0000000000000246 ffff8ae9e4857040
Apr 12 14:48:12 alan kernel: Call Trace:
Apr 12 14:48:12 alan kernel:  [<ffffffff973f6403>] ? __schedule+0x233/0x6d0
Apr 12 14:48:12 alan kernel:  [<ffffffff973f68d2>] ? schedule+0x32/0x80
Apr 12 14:48:12 alan kernel:  [<ffffffff973f9ca3>] ? schedule_timeout
+0x243/0x310
Apr 12 14:48:12 alan kernel:  [<ffffffff96ee52f6>] ? add_timer+0x126/0x200
Apr 12 14:48:12 alan kernel:  [<ffffffff973f72e1>] ? wait_for_completion
+0xf1/0x130
Apr 12 14:48:12 alan kernel:  [<ffffffff96ea17f0>] ? wake_up_q+0x70/0x70
Apr 12 14:48:12 alan kernel:  [<ffffffffc0401afe>] ? xhci_setup_device+0x19e/
0x730 [xhci_hcd]
Apr 12 14:48:12 alan kernel:  [<ffffffffc0387e21>] ? hub_port_init+0x181/0xba0
[usbcore]
Apr 12 14:48:12 alan kernel:  [<ffffffffc038c1d3>] ? hub_event+0xb43/0x14d0
[usbcore]
Apr 12 14:48:12 alan kernel:  [<ffffffff96e90384>] ? process_one_work
+0x184/0x410
Apr 12 14:48:12 alan kernel:  [<ffffffff96e9065d>] ? worker_thread+0x4d/0x480
Apr 12 14:48:12 alan kernel:  [<ffffffff96e90610>] ? process_one_work
+0x410/0x410
Apr 12 14:48:12 alan kernel:  [<ffffffff96e90610>] ? process_one_work
+0x410/0x410
Apr 12 14:48:12 alan kernel:  [<ffffffff96e965ce>] ? kthread+0xce/0xf0
Apr 12 14:48:12 alan kernel:  [<ffffffff96e24701>] ? __switch_to+0x2c1/0x6c0
Apr 12 14:48:12 alan kernel:  [<ffffffff96e96500>] ? kthread_park+0x60/0x60
Apr 12 14:48:12 alan kernel:  [<ffffffff973fb2f5>] ? ret_from_fork+0x25/0x30





Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Jochen Spieker
solitone:
>
> Hi, something really weird is going on today on my system--debian stretch on
> an Apple MacBookPro 12,1.

If googling doesn't turn up something helpful (and maybe even then) I'd
write a bug report. Your e-mail is a pretty good start.

Regards,
Jochen.
--
If politics is the blind leading the blind, entertainment is the fucked-
up leading the hypnotised.
[Agree]   [Disagree]
                 <http://archive.slowlydownward.com/NODATA/data_enter2.html>

signature.asc (836 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
On Wednesday, 12 April 2017 19:55:12 CEST Jochen Spieker wrote:
> I'd write a bug report. Your e-mail is a pretty good start.

To Debian BTS? Related to the kernel package? I have no clues as to what
component might be actually involved.

Thanks,
  Davide


Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Mart van de Wege
solitone <[hidden email]> writes:

> On Wednesday, 12 April 2017 19:55:12 CEST Jochen Spieker wrote:
>> I'd write a bug report. Your e-mail is a pretty good start.
>
> To Debian BTS? Related to the kernel package? I have no clues as to what
> component might be actually involved.
>
It's USB-related, so I'd say either the kernel package or udev.

Mart

--
"We will need a longer wall when the revolution comes."
    --- AJS, quoting an uncertain source.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Jochen Spieker
In reply to this post by solitone
solitone:
> On Wednesday, 12 April 2017 19:55:12 CEST Jochen Spieker wrote:
>> I'd write a bug report. Your e-mail is a pretty good start.
>
> To Debian BTS? Related to the kernel package? I have no clues as to what
> component might be actually involved.

I'd use reportbug againt the kernel package.

J.
--
I think of my genitals more often than my hands, but use them far less.
[Agree]   [Disagree]
                 <http://archive.slowlydownward.com/NODATA/data_enter2.html>

signature.asc (836 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Michael Lange-2
In reply to this post by solitone
On Wed, 12 Apr 2017 22:13:44 +0200
solitone <[hidden email]> wrote:

> On Wednesday, 12 April 2017 19:55:12 CEST Jochen Spieker wrote:
> > I'd write a bug report. Your e-mail is a pretty good start.
>
> To Debian BTS? Related to the kernel package? I have no clues as to
> what component might be actually involved.
>
> Thanks,
>   Davide
>
>
>

I would try to install a different kernel (if possible with external
keyboard) and boot into that one, if the problem disappears the culprit is
most likely the kernel.

Regards

Michael

.-.. .. ...- .   .-.. --- -. --.   .- -. -..   .--. .-. --- ... .--. . .-.

You say you are lying.  But if everything you say is a lie, then you are
telling the truth.  You cannot tell the truth because everything you say
is a lie.  You lie, you tell the truth ... but you cannot, for you lie.
                -- Norman the android, "I, Mudd", stardate 4513.3

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Dejan Jocic
On 13-04-17, Michael Lange wrote:

> On Wed, 12 Apr 2017 22:13:44 +0200
> solitone <[hidden email]> wrote:
>
> > On Wednesday, 12 April 2017 19:55:12 CEST Jochen Spieker wrote:
> > > I'd write a bug report. Your e-mail is a pretty good start.
> >
> > To Debian BTS? Related to the kernel package? I have no clues as to
> > what component might be actually involved.
> >
> > Thanks,
> >   Davide
> >
> >
> >
>
> I would try to install a different kernel (if possible with external
> keyboard) and boot into that one, if the problem disappears the culprit is
> most likely the kernel.
>
He should have old kernel still installed, right? If that is the case,
he could simply boot with old kernel.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
In reply to this post by Mart van de Wege
On Wednesday, 12 April 2017 22:29:25 CEST Mart van de Wege wrote:
> It's USB-related, so I'd say either the kernel package or udev.

Ok, thanks. These are the related packages that were upgrated:

linux-image-4.9.0-2-amd64:amd64 (4.9.13-1, 4.9.18-1)
udev:amd64 (232-19, 232-22)
libudev1:amd64 (232-19, 232-22)


Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
In reply to this post by Dejan Jocic
On Thursday, 13 April 2017 01:23:02 CEST Dejan Jocic wrote:
> He should have old kernel still installed, right? If that is the case,
> he could simply boot with old kernel.

Yes, I still have 4.9.0-1-amd64.

The point is even in the GRUB menu screen my keyboard no longer works well,
which is even stranger. Sometimes I press a key and after tens of seconds
something happens, other times nothing happens at all. So it's difficult to move
in the GRUB menu and select the old kernel.

Plus now it's been 18 hours since last boot and everything works very well,
but if I reboot now I'll struggle again and I'm not sure I'll manage to have
it work again..

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Dejan Jocic
On 13-04-17, solitone wrote:

> On Thursday, 13 April 2017 01:23:02 CEST Dejan Jocic wrote:
> > He should have old kernel still installed, right? If that is the case,
> > he could simply boot with old kernel.
>
> Yes, I still have 4.9.0-1-amd64.
>
> The point is even in the GRUB menu screen my keyboard no longer works well,
> which is even stranger. Sometimes I press a key and after tens of seconds
> something happens, other times nothing happens at all. So it's difficult to move
> in the GRUB menu and select the old kernel.
>
> Plus now it's been 18 hours since last boot and everything works very well,
> but if I reboot now I'll struggle again and I'm not sure I'll manage to have
> it work again..
>

If you have old kernel, you do not have to choose it in the GRUB menu during
boot, you can set up your GRUB to boot from it automatically. Glad to
hear that it works well now, though you will probably have to reboot
again, sooner or later.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
On Thursday, 13 April 2017 09:53:05 CEST Dejan Jocic wrote:
> If you have old kernel, you do not have to choose it in the GRUB menu during
> boot, you can set up your GRUB to boot from it automatically.

I have a submenu entry in my grub.cfg:

solitone@alan:~$ grep --color menu /boot/grub/grub.cfg                                                                              
[...]
menuentry 'Debian GNU/Linux' --class debian --class gnu-linux --class gnu --
class os $menuentry_id_option 'gnulinux-simple-7f537d3b-e578-4cd3-8583-
f3a2281f7d01' {                                                                                                        
submenu 'Advanced options for Debian GNU/Linux' $menuentry_id_option
'gnulinux-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64' --class debian
--class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-2-amd64-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {                                                      
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64 (systemd)' --
class debian --class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-2-amd64-init-systemd-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64 (recovery mode)'
--class debian --class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-2-amd64-recovery-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64' --class debian
--class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-1-amd64-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64 (systemd)' --
class debian --class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-1-amd64-init-systemd-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
        menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64 (recovery mode)'
--class debian --class gnu-linux --class gnu --class os $menuentry_id_option
'gnulinux-4.9.0-1-amd64-recovery-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {

To automatically load 'Debian GNU/Linux, with Linux 4.9.0-1-amd64' I need to
set :
GRUB_DEFAULT="1>3"
in /etc/default/grub?

Cheers,
  Davide
 

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
In reply to this post by Michael Lange-2
On Thursday, 13 April 2017 00:27:22 CEST Michael Lange wrote:
> I would try to install a different kernel (if possible with external
> keyboard) and boot into that one, if the problem disappears the culprit is
> most likely the kernel.

I managed to boot into the older kernel that I still had in the grub menu
(4.9.0-1-amd64). It was difficult because, as I said, the keyboard works badly
in GRUB as well, but at least it somehow works, I just have to wait tens of
seconds after a keypress.

Anyhow, it booted in 4.9.0-1-amd64, but it showed the same issue: keyboard and
trackpad don't work. So it doesn't seem to be kernel related.

Now I have rebooted again in 4.9.0-2-amd64 version 4.9.18-1 (2017-03-30).
Apparenlty the only way I can restore functioning of the integrated keyboard
and trackpad is by booting up with an external usb keyboard and an external
usb mouse plugged in. I have the usual issues in grub and during bootup, but
when I get the login screen, after 30 seconds or so the external usb devices
get detected, and also the integrated keyboard and trackpad get detected so
everything starts working.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Dejan Jocic
In reply to this post by solitone
On 13-04-17, solitone wrote:

> On Thursday, 13 April 2017 09:53:05 CEST Dejan Jocic wrote:
> > If you have old kernel, you do not have to choose it in the GRUB menu during
> > boot, you can set up your GRUB to boot from it automatically.
>
> I have a submenu entry in my grub.cfg:
>
> solitone@alan:~$ grep --color menu /boot/grub/grub.cfg                                                                              
> [...]
> menuentry 'Debian GNU/Linux' --class debian --class gnu-linux --class gnu --
> class os $menuentry_id_option 'gnulinux-simple-7f537d3b-e578-4cd3-8583-
> f3a2281f7d01' {                                                                                                        
> submenu 'Advanced options for Debian GNU/Linux' $menuentry_id_option
> 'gnulinux-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64' --class debian
> --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-2-amd64-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {                                                      
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64 (systemd)' --
> class debian --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-2-amd64-init-systemd-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-2-amd64 (recovery mode)'
> --class debian --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-2-amd64-recovery-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64' --class debian
> --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-1-amd64-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64 (systemd)' --
> class debian --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-1-amd64-init-systemd-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>         menuentry 'Debian GNU/Linux, with Linux 4.9.0-1-amd64 (recovery mode)'
> --class debian --class gnu-linux --class gnu --class os $menuentry_id_option
> 'gnulinux-4.9.0-1-amd64-recovery-7f537d3b-e578-4cd3-8583-f3a2281f7d01' {
>
> To automatically load 'Debian GNU/Linux, with Linux 4.9.0-1-amd64' I need to
> set :
> GRUB_DEFAULT="1>3"
> in /etc/default/grub?
>
> Cheers,
>   Davide
>  
You can set it with numbers, but you can set it also with quotes, like
for example:

GRUB_DEFAULT='gnulinux-4.9.0-1-amd64-advanced-7f537d3b-e578-4cd3-8583-f3a2281f7d01'

Which should be kind of more secure way if you do not to want to
miss what you exactly want with numbers.

Anyway, however that you do it, after that you will need to run
# update-grub.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

Michael Lange-2
In reply to this post by solitone
On Thu, 13 Apr 2017 11:02:06 +0200
solitone <[hidden email]> wrote:

> On Thursday, 13 April 2017 00:27:22 CEST Michael Lange wrote:
> > I would try to install a different kernel (if possible with external
> > keyboard) and boot into that one, if the problem disappears the
> > culprit is most likely the kernel.
>
> I managed to boot into the older kernel that I still had in the grub
> menu (4.9.0-1-amd64). It was difficult because, as I said, the keyboard
> works badly in GRUB as well, but at least it somehow works, I just have
> to wait tens of seconds after a keypress.
>
> Anyhow, it booted in 4.9.0-1-amd64, but it showed the same issue:
> keyboard and trackpad don't work. So it doesn't seem to be kernel
> related.
>
> Now I have rebooted again in 4.9.0-2-amd64 version 4.9.18-1
> (2017-03-30). Apparenlty the only way I can restore functioning of the
> integrated keyboard and trackpad is by booting up with an external usb
> keyboard and an external usb mouse plugged in. I have the usual issues
> in grub and during bootup, but when I get the login screen, after 30
> seconds or so the external usb devices get detected, and also the
> integrated keyboard and trackpad get detected so everything starts
> working.
>

Another thought: if that is possible, have you tried to boot into a live
system, just to rule out a hardware issue?

Regards

Michael


.-.. .. ...- .   .-.. --- -. --.   .- -. -..   .--. .-. --- ... .--. . .-.

Spock: The odds of surviving another attack are 13562190123 to 1, Captain.

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
On Thursday, 13 April 2017 11:20:34 CEST Michael Lange wrote:
> Another thought: if that is possible, have you tried to boot into a live
> system, just to rule out a hardware issue?

I had used Apple Diagnostics to perform a hardware check, and nothing wrong
was found.

After that, I tried and reset NVRAM, a nonvolatile random-access memory that
Macs use to store certain settings like sound volume, display resolution, and
(I realise only now) startup disk selection.. and GRUB doesn't start any
longer, it boots up directly into MacOS.

After that, I also reset the System Management Controller (SMC) and at least
now the initial bootup phase into MacOS is much quicker.

Now I managed to restore GRUB, using the rescue mode option of my installation
USB stick. Everything is as before: very slow keystrokes response in grub,
integrated keyboard and mouse working after some time in login screen only
when external usb mouse and keyboard plugged in (both of them!)

I've noticed that also Apple Startup Manager suffers from the same issue that
GRUB has: I press keys and it responds after tens of seconds! Trackpad is also
very slow or doesn't work at all. I need to check my updates on Apple side..

Reply | Threaded
Open this post in threaded view
|

Re: System broken after yesterday's upgrade

solitone
In reply to this post by solitone
On Thursday, 13 April 2017 11:02:06 CEST solitone wrote:
> Apparenlty the only way I can restore functioning of the integrated keyboard
> and trackpad is by booting up with an external usb keyboard and an external
> usb mouse plugged in. I have the usual issues in grub and during bootup,
> but when I get the login screen, after 30 seconds or so the external usb
> devices get detected, and also the integrated keyboard and trackpad get
> detected so everything starts working.

This is what I collect from a ssh, running multiple lsusb commands.

At first only the two external devices are listed, apart from the hubs:

solitone@alan:~$ lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 003: ID 04b3:3025 IBM Corp. NetVista Full Width Keyboard
Bus 001 Device 002: ID 046d:c05a Logitech, Inc. M90/M100 Optical Mouse
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

Then one of the two integrated usb devices (maybe the keyboard, I haven't
checked, so I don't know for sure) is detected:

solitone@alan:~$ lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 05ac:8290 Apple, Inc.
Bus 001 Device 003: ID 04b3:3025 IBM Corp. NetVista Full Width Keyboard
Bus 001 Device 002: ID 046d:c05a Logitech, Inc. M90/M100 Optical Mouse
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

After some further time, also the second integrated usb device (perhaps the
trackpad) is detected:

solitone@alan:~$ lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 005: ID 05ac:0273 Apple, Inc.
Bus 001 Device 004: ID 05ac:8290 Apple, Inc.
Bus 001 Device 003: ID 04b3:3025 IBM Corp. NetVista Full Width Keyboard
Bus 001 Device 002: ID 046d:c05a Logitech, Inc. M90/M100 Optical Mouse
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

>From this time, everything works fine, and I can also unplug the external usb
devices and the integrated keyboard and trackpad continue to work.

Cheers,
  Davide



Reply | Threaded
Open this post in threaded view
|

Re: [SOLVED] System broken after yesterday's upgrade

solitone
In reply to this post by solitone
On Thursday, 13 April 2017 12:44:06 CEST solitone wrote:

>  I tried and reset NVRAM, a nonvolatile random-access memory that
> Macs use to store certain settings like sound volume, display resolution,
> and (I realise only now) startup disk selection.. and GRUB doesn't start
> any longer, it boots up directly into MacOS.
>
> After that, I also reset the System Management Controller (SMC) and at least
> now the initial bootup phase into MacOS is much quicker.
>
> Now I managed to restore GRUB, using the rescue mode option of my
> installation USB stick. Everything is as before: very slow keystrokes
> response in grub, integrated keyboard and mouse working after some time in
> login screen only when external usb mouse and keyboard plugged in (both of
> them!)
>
> I've noticed that also Apple Startup Manager suffers from the same issue
> that GRUB has: I press keys and it responds after tens of seconds! Trackpad
> is also very slow or doesn't work at all. I need to check my updates on
> Apple side..

The thing I didn't explain there is that after I reset the SMC, MacOS boot
seemed to be quicker, however it still tooked 30 seconds or more to start the
Startup Manager when pressing the alt button at startup. Besides, Startup
Manager responded very slowly to keypresses, and the trackpad didn't work, or
was really slow. This clarifies the issue wasn't a consequence of reinstalling
GRUB, and I suspected it depended either on Apple's hardware or in some low
level Apple software component.

I found a few posts around that described exactly the same problem I had, like
this one:
https://apple.stackexchange.com/questions/270248/macos-startup-manager-is-very-slow-and-laggy

Some people said that MacOS Siera 10.12.4 would solve that issue. So I tried
and upgraded (I was on 10.12.3), and I can confirm it is indeed true. That
system update must contain also a firmware update that addresses that I/O
issue. Now everything works very well again.