Bug in laptop-mode-tools-1.65? [Solved]

Fitzcarraldo · Last edited by Fitzcarraldo on Fri Oct 24, 2014 8:10 am; edited 1 time in total

I have just completed an emerge -uvDN world. I notice that upower-pm-utils was upgraded from 0.9.23-r1 to 0.9.23-r2 and laptop-mode-tools was upgraded from 1.64-r2 to 1.65:

Fitzcarraldo · Posted: Wed Jun 25, 2014 7:34 pm Post subject:

I have done some investigating, and I'm beginning to suspect there is a bug in laptop-mode-tools version 1.65.

First I downgraded laptop-mode-tools back to version 1.64-r2:

Fitzcarraldo · Posted: Tue Oct 21, 2014 6:16 pm Post subject:

Well, after my post on 25 June I masked >=laptop-mode-tools-1.65, merged 1.64-r2 again and things were back to normal. Then in August I saw Gentoo Bugzilla Bug Report No. 520124 ('app-laptop/laptop-mode-tools-1.65 has broken usb-autosuspend modules') and all became clear(er) thanks to fellow Gentoo and laptop-mode-tools user Ted Tanberry.

A week ago I noticed that an ebuild for laptop-mode-tools-1.66 is available in the Portage main tree and decided to try it. The results were rather bizarre:

a) Sometimes, after booting everything works as expected: the mouse and keyboard do not auto-suspend, and the KDE network management widget shows an available wired connection named "Wired connection 1" (which is eth0).

b) However, sometimes, after booting the mouse and keyboard auto-suspend and the KDE network management widget shows an extra available connection "eth0" in addition to "Wired connection 1". Furthermore, I can connect to "eth0" or "Wired connection 1", but neither provides access to the Internet.

c) In the first case mentioned above, if I then Suspend to RAM and wake the laptop from suspension, from then onwards the keyboard and mouse auto-suspend.

Neither Case b nor Case c above occurred with laptop-mode-tools-1.64.

I despair sometimes! :roll:

EDIT: Actually, I have now discovered that in Case b sometimes it is possible to access the Internet by selecting either "eth0" or "Wired connection 1".
_________________
Clevo W230SS: amd64, VIDEO_CARDS="intel modesetting nvidia".
Compal NBLB2: ~amd64, xf86-video-ati. Dual boot Win 7 Pro 64-bit.
OpenRC udev elogind & KDE on both.
Fitzcarraldo's blog

Roman_Gruber · Posted: Wed Oct 22, 2014 9:42 am Post subject:

usb-autosuspend seems the cause for your issues? maybe worth getting rid in the kernel or unloading the module always?

just an idea.

Fitzcarraldo · Posted: Wed Oct 22, 2014 11:55 am Post subject:

Apparently CONFIG_USB_SUSPEND was dropped after kernel 3.9:

http://cateee.net/lkddb/web-lkddb/USB_SUSPEND.html

Fitzcarraldo · Posted: Fri Oct 24, 2014 8:10 am Post subject:

OK, it seems that LMT 1.66 is working as intended. Mind you, it would probably be better to do what Ted Tanberry recommended in Gentoo Bugzilla Bug Report No. 520124, i.e. edit /etc/laptop-mode/conf.d/runtime-pm.conf as follows, to stop all external keyboards, mice and USB HDDs going to sleep if the laptop is on battery power:

kernelOfTruth · Posted: Sat Oct 25, 2014 8:14 pm Post subject:

Hi Fitzcarraldo,

I'm linking to a post which links back to your blog - hope it helps somehow:

https://forums.gentoo.org/viewtopic-t-995446-highlight-.html
_________________
https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.9.0
https://github.com/kernelOfTruth/pulseaudio-equalizer-ladspa

Hardcore Gentoo Linux user since 2004

Fitzcarraldo · Posted: Sun Oct 26, 2014 4:38 pm Post subject:

Thanks for posting kernelOfTruth but, as mentioned in my previous post, I sorted out my problem with LMT 1.66. It's a pity I did not see the thread you referenced, as it took me a while to understand precisely what Gentoo Bugzilla Bug Report No. 520124 meant, i.e. that LMT 1.66 had switched to using LMT module runtime-pm instead of LMT module usb-autosuspend.

In the thread you referenced, Xywa mentioned a previous alternative approach of disabling USB auto-suspend, by using kernel boot line parameter 'usbcore.autosuspend=-1' (or 'options usbcore autosuspend=-1' in /etc/modprobe.d/usbcore.conf if usbcore is built as a kernel module instead of in-kernel), but, as we know, CONFIG_USB_SUSPEND was dropped after kernel 3.9. Hence LMT is the way to go.

I now think the extra network connection ('eth0') appearing in the list of available connections in my KDE Panel may be due to net-misc/networkmanager, as that extra entry 'eth0' started appearing in the KDE Plasma NM widget's list of available connections after a world update I performed on 13 October merged net-misc/networkmanager-0.9.10.0. Anyway, today I used the KDE Plasma NM widget's connection editor to delete the phantom 'eth0' from its list of available connections to see what would happen, and the next time I booted the laptop the wired connection would no longer work. So I used the KDE Plasma NM widget's connection editor to manually add a new wired connection which I named 'eth0', and the wired connection 'Wired connection 1' now works again. :roll:

So I'll leave it.

EDIT Mon 10 Nov 22:13:02 GMT 2014: I think I may have found the reason for the extra entry ("eth0") in the list of available connections in the plasma-nm widget on my KDE Panel. I have now deleted the file /etc/udev/rules.d/70-persistent-net.rules and initscript /etc/init.d/rename_ethX (given in Gentoo Bug Report No. 453494) as I think they were obsoleted by the following two files in later releases of udev for those of us who want to maintain the kernel naming scheme "eth0" and "wlan0" instead of the systemd/udev 'predictable network interface names':

chutzpah · Posted: Tue Nov 04, 2014 5:49 am Post subject:

These udev rules will enable autosuspend for all USB devices except mice and keyboards (you can turn off laptop-mode's support for this after these are added)

Fitzcarraldo · Posted: Sun Feb 15, 2015 6:28 pm Post subject:

Just in case someone finds this thread by searching, I want to clarify that the problem of the additional connection entry 'eth0' in the plasma-nm widget was caused by NetworkManager, not udev.

Although I wrote in my two edits of my previous post that deleting some udev files and adding the kernel boot parameter 'net.ifnames=0' appeared to have eliminated the unwanted extra connection 'eth0' in the NetworkManager GUI front-end, I was mistaken. (Mind you, deleting those udev files and adding kernel boot parameter 'net.ifnames=0' did reliably rid me of the so-called Predictable Network Interface Names, so that was grist to the mill.)

I have been breaking my head over this for months and tracked it down to NetworkManager. The big breakthrough yesterday was finding Debian bug report no. 755202: network-manager: keeps creating and using new connection “eth0″ that does not work. My blog post today NetworkManager creates a new connection ‘eth0′ that does not work goes into detail, but the bottom line is to do the following if you find an extra connection appearing for no good reason:

1. Check that your network driver is reliable. You can search the Web to see if other users have experienced problems with the same driver you are using.

2. Make sure the contents of NetworkManager.conf are correct. (See the NetworkManager.conf man page and the GNOME Wiki page on NetworkManager settings to find out what options are available.)

3. Delete all the files (including hidden files) in the directory /etc/NetworkManager/system-connections/ and recreate your connections via the NetworkManager GUI (plasma-nm, nm-applet or whatever) or TUI (nmtui).
_________________
Clevo W230SS: amd64, VIDEO_CARDS="intel modesetting nvidia".
Compal NBLB2: ~amd64, xf86-video-ati. Dual boot Win 7 Pro 64-bit.
OpenRC udev elogind & KDE on both.
Fitzcarraldo's blog

Fitzcarraldo · Posted: Tue Mar 10, 2015 1:41 pm Post subject:

Oh, man. I thought the 'additional eth0 entry' problem was solved, but it isn't. :evil:

My previous post was made while I was away from home on a work trip, and the changes I described seemed to have cured the problem on the networks my laptop was connected to (wired and wireless networks in both office and hotel). I am now back at home and my laptop is connected to my home network via an Ethernet cable. Even with the changes I made (see my previous post), when I boot the laptop NetworkManager sometimes (but not always) has two connections named ‘eth0′, one of them the ‘Active connection’ (but not able to connect to the Internet) and the other an ‘Available connection’. In this situation the wired network icon on the Panel has a yellow question mark superimposed. If I delete the ‘eth0′ active connection and use the other ‘eth0′, the latter works as expected and I have no trouble connecting to the Internet. In Debian bug report no. 755202 user Frederik Himpe added a comment on March 4, 2015 that he is also still experiencing this problem and “It looks like there is a race somewhere, causing the network interface to be brought up before Network Manager is started, and this prevents correct configuration by NM”. So the problem is still unresolved.

Hmm … I wonder if udev does have something to do with it after all.

Is anyone else experiencing this behaviour?
_________________
Clevo W230SS: amd64, VIDEO_CARDS="intel modesetting nvidia".
Compal NBLB2: ~amd64, xf86-video-ati. Dual boot Win 7 Pro 64-bit.
OpenRC udev elogind & KDE on both.
Fitzcarraldo's blog