hoibn
n00b
Joined: 28 May 2014
Posts: 2
|
 Posted: Wed May 28, 2014 9:51 pm Post subject: Reconnecting ethernet cable causes a shutdown |
 |
|
More specifically, a acpid shutdown, as if I had pressed power down button. Removing the ethernet cable, waiting ~5 seconds, then reconnecting. Too little time and it doesn't work. And it sometimes doesn't work even if I wait 10 seconds. Happens on 3.15-rc7, 3.14.3 and 3.12.13
Set acpid to not start manually then started it with --debug on a console, then pressed power down button once then did the disconnect reconnect cable (I changed acpid to not shutdown on power button press naturally):
| Quote: |
* Starting acpid ...Deprecated /proc/acpi/event was not found. Trying netlink and the input layer...
input layer /dev/input/event0 (Video Bus) opened successfully, fd 4
input layer /dev/input/event1 (AT Translated Set 2 keyboard) opened successfully, fd 5
input layer /dev/input/event10 (HDA NVidia HDMI/DP,pcm=3) opened successfully, fd 6
input layer /dev/input/event11 (HDA NVidia HDMI/DP,pcm=7) opened successfully, fd 7
input layer /dev/input/event12 (HDA NVidia HDMI/DP) opened successfully, fd 8
input layer /dev/input/event13 (HDA Intel MID Headphone) opened successfully, fd 9
input layer /dev/input/event4 (Logitech USB Receiver) opened successfully, fd 10
input layer /dev/input/event5 (Power Button) opened successfully, fd 11
input layer /dev/input/event6 (Lid Switch) opened successfully, fd 12
input layer /dev/input/event7 (Sleep Button) opened successfully, fd 13
input layer /dev/input/event8 (Power Button) opened successfully, fd 14
inotify fd: 15
inotify wd: 1
netlink opened successfully
acpid: starting up with netlink and the input layer
parsing conf file /etc/acpi/events/default
parsing conf file /etc/acpi/events/lm_ac_adapter
parsing conf file /etc/acpi/events/lm_lid
parsing conf file /etc/acpi/events/lm_battery
acpid: 4 rules loaded
acpid: waiting for events: event logging is on
acpid: received input layer event "button/power PBTN 00000080 00000000"
acpid: rule from /etc/acpi/events/default matched
acpid: executing action "/etc/acpi/default.sh button/power PBTN 00000080 00000000"
acpid: action exited with status 0
acpid: 1 total rule matched
acpid: completed input layer event "button/power PBTN 00000080 00000000"
acpid: received netlink event "button/power LNXPWRBN:00 00000080 00000003"
acpid: rule from /etc/acpi/events/default matched
acpid: executing action "/etc/acpi/default.sh button/power LNXPWRBN:00 00000080 00000003"
acpid: action exited with status 0
acpid: 1 total rule matched
acpid: completed netlink event "button/power LNXPWRBN:00 00000080 00000003"
acpid: received input layer event "button/power PBTN 00000080 00000000"
acpid: rule from /etc/acpi/events/default matched
acpid: executing action "/etc/acpi/default.sh button/power PBTN 00000080 00000000"
acpid: action exited with status 0
acpid: 1 total rule matched
acpid: completed input layer event "button/power PBTN 00000080 00000000"
acpid: received netlink event "button/power LNXPWRBN:00 00000080 00000004"
acpid: rule from /etc/acpi/events/default matched
acpid: executing action "/etc/acpi/default.sh button/power LNXPWRBN:00 00000080 00000004"
acpid: action exited with status 0
acpid: 1 total rule matched
acpid: completed netlink event "button/power LNXPWRBN:00 00000080 00000004"
|
By running (is there a better way?)
$ cat /dev/input/event8
I can check that it's event8 that's causing the problem, as cat shows something both if I press power button and reconnect the cable.
$ udevadm info --query=all --name=/dev/input/event8
| Quote: |
P: /devices/LNXSYSTM:00/LNXPWRBN:00/input/input9/event8
N: input/event8
E: DEVNAME=/dev/input/event8
E: DEVPATH=/devices/LNXSYSTM:00/LNXPWRBN:00/input/input9/event8
E: ID_INPUT=1
E: ID_INPUT_KEY=1
E: ID_PATH=acpi-LNXPWRBN:00
E: ID_PATH_TAG=acpi-LNXPWRBN_00
E: MAJOR=13
E: MINOR=72
E: SUBSYSTEM=input
E: USEC_INITIALIZED=469391
|
Unloading either the r8169 module (my ethernet card module) or the (ACPI) button module avoids the problem, but breaks something else. Reloading button restores the problem. Reloading r8169 requires restarting net.enp3s0 to get the network back. Restarting net.enp3s0 restores the problem.
evbug shows essentially the same thing both for a reconnect and a power button press.
For a power button press
| Quote: | evbug: Event. Dev: input18, Type: 1, Code: 116, Value: 1
evbug: Event. Dev: input18, Type: 0, Code: 0, Value: 0
evbug: Event. Dev: input18, Type: 1, Code: 116, Value: 0
evbug: Event. Dev: input18, Type: 0, Code: 0, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 1000
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 1000
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 1000
evbug: Event. Dev: input10, Type: 0, Code: 0, Value: 1
|
And for a reconnect
| Quote: |
evbug: Event. Dev: input18, Type: 1, Code: 116, Value: 1
evbug: Event. Dev: input18, Type: 0, Code: 0, Value: 0
evbug: Event. Dev: input18, Type: 1, Code: 116, Value: 0
evbug: Event. Dev: input18, Type: 0, Code: 0, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 1000
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 1000
evbug: Event. Dev: input10, Type: 18, Code: 2, Value: 0
evbug: Event. Dev: input10, Type: 18, Code: 1, Value: 0
evbug: Event. Dev: input10, Type: 0, Code: 0, Value: 1
|
the problem doesn't exist in archlinux with 3.12.13, so it's a possible error while manually configuring the kernel, or something else. I used gentoo-sources, except for the rc kernel (which was used for the logs), which is git-sources. Rc kernel is blobed and the others ones are deblobed. Is the problem on kernel or userspace by the way? Also, is there a weirdest problem prize  ? |
|
Kernel & Hardware
