LinuxQuestions.org
Download your favorite Linux distribution at LQ ISO.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Newbie
User Name
Password
Linux - Newbie This Linux forum is for members that are new to Linux.
Just starting out and have a question? If it is not in the man pages or the how-to's this is the place!

Notices


Reply
  Search this Thread
Old 02-11-2011, 10:02 PM   #1
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Rep: Reputation: 0
Debian 686 kernel total networking failure on ibmX60, but 486 kernel works


The 486 kernel works just fine, and while I have only 1GB of RAM at the moment I hope to have 2GB someday and would like to take advantage of the dual core CPU, so I would like to configure grub to run the 686 kernel by default. For whatever reason, it runs the 486 right now and the 686 fails in a major way: there is no network connectivity at all. It could be plugged into my cable modem router and it shows no wired connections. The fact that one works and the other doesn't puzzles me since I haven't touched either since the install and a few rounds of upgrades. The fact that this is an X60 makes everything a little esoteric, and harder for a newbie.

I should mention I'm newbie but getting better; I managed to install debian on this x60, yet while preserving the factory install rescue & recovery partition and preserving the factory install MBR so that ibm-specific hardware functions (thinkvantage button, etc.) still work. This required me to use dd to copy the first 512 bytes of my debian partition to a file in the windows partition, etc., and modifying the windows bootloader. (I wish I had learned dd long ago--it rocks). I did this because if I ever resell the X60, the fact is most people use MS Windows and having that partition adds a perception of value to some potential buyers; not to mention I paid $ for it (I was young & stupid) so why should I delete it. I also backed up the recovery partition on another drive using dd over NFS in case the hd ever heads south.


Anyway, I've never been comfy with messing with the kernel. I did once recompile a module for ALSA because it had a bug in it for an old Yamaha integrated sound card on an old PIII and the newer version worked [alsa fails on this x60 too but I think I found a post on here that has a solution I will try later]. But I'm clueless as to networking modules, not to mention the correct module is installed already from Intel for this chipset. So what is there to do?

Here's a clue: the ifconfig output is radically different from the 686 and 486 kernels. Looks like hardware is not being detected since eth0 fails to show:
================686 ifconfig output====================
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:70 errors:0 dropped:0 overruns:0 frame:0
TX packets:70 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:5312 (5.1 KiB) TX bytes:5312 (5.1 KiB)

====================486 ifconfig output=============================

eth0 Link encap:Ethernet HWaddr 00:16:d3:23:67:f0
inet addr:192.168.1.27 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::216:d3ff:fe23:67f0/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:166 errors:0 dropped:0 overruns:0 frame:0
TX packets:58 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:11071 (10.8 KiB) TX bytes:7936 (7.7 KiB)
Memory:ee000000-ee020000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:108 errors:0 dropped:0 overruns:0 frame:0
TX packets:108 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:8888 (8.6 KiB) TX bytes:8888 (8.6 KiB)

I would show the diff output below if it weren't so long--and not allowed--upon 2 text files, the first holding the output of modprobe -l under the 486 kernel and the second under the 686 kernel. I could pm it. Any suggestions or clues where to start would be very welcome.
 
Old 02-12-2011, 02:22 PM   #2
business_kid
LQ Guru
 
Registered: Jan 2006
Location: Ireland
Distribution: Slackware & Android
Posts: 7,565

Rep: Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696
uname -r on both kernels - that's where to start. If they are the same, that's a problem.
/lib/modules/kernel-version is where your modules go. Try ls -ld /lib/modules/*
Also, think through any initrd you use. Is _that_ built with '486 or '686 modules?

Last edited by business_kid; 02-12-2011 at 02:23 PM.
 
Old 02-13-2011, 12:43 PM   #3
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
making progress...?

Here's the output of uname with everything:
Linux x60 2.6.32-5-486 #1 Wed Jan 12 03:28:50 UTC 2011 i686 GNU/Linux
Linux x60 2.6.26-2-686 #1 SMP Thu Jan 27 00:28:05 UTC 2011 i686 GNU/Linux
So they're not the same...so that might be good as you said?

Modules as per ls -ld /lib/modules/*:
drwxr-xr-x 3 root root 4096 Feb 6 01:02 /lib/modules/2.6.26-2-686
drwxr-xr-x 4 root root 4096 Feb 7 02:50 /lib/modules/2.6.32-5-486
drwxr-xr-x 3 root root 4096 Feb 7 02:51 /lib/modules/2.6.32-5-686

I did some reading about what an initrd is to begin with, then how to uncompress it and extract its contents with cpio: http://www.ibm.com/developerworks/li.../l-initrd.html, and the first link from there, and http://www.gnu.org/software/cpio/man...mono/cpio.html. But I am at a loss to "think through" any initrd I use and if it is built with certain modules or others.

How can I determine that since I did not build any initrd manually, but rather it was created by an automated process during the initial install and any subsequent upgrades. How do I determine which modules are 486 versus 686 modules? I presume if it's not in the name there's a list somewhere? Does a record of that build process exist somewhere? Or is it sufficient to examine the contents of the initrd image itself?

I decompressed the initrd images for the 486 and 686 kernels, and using the -v option with cpio, and then ran the output through diff, which is below [I search replaced the kernel versions with "--" otherwise every single line with a path that includes a kernel version would be different even if the terminal file name was identical]. Obviously there there have to be big differences between these 2 kernels, but the devil is in finding which are correct and which are causing problems. Thanks for your help.

====486 initrd==== ====686 initrd=====
. .
init init
etc etc
etc/udev etc/udev
etc/udev/udev.conf etc/udev/udev.conf
> etc/udev/links.conf
> etc/udev/.dev
> etc/udev/libgphoto2.rules
> etc/udev/libmtp7.rules
> etc/udev/alsa-utils.rules
> etc/udev/rules.d
> etc/udev/rules.d/z60_xserver-xorg-input-wacom.rules
> etc/udev/rules.d/60-persistent-storage-tape.rules
> etc/udev/rules.d/75-persistent-net-generator.rules
> etc/udev/rules.d/91-permissions.rules
> etc/udev/rules.d/70-persistent-cd.rules
> etc/udev/rules.d/70-persistent-net.rules
> etc/udev/rules.d/95-late.rules
> etc/udev/rules.d/50-udev.rules
> etc/udev/rules.d/libmtp7.rules
> etc/udev/rules.d/60-persistent-storage.rules
> etc/udev/rules.d/90-hal.rules
> etc/udev/rules.d/85-pcmcia.rules
> etc/udev/rules.d/z60_libsane.rules
> etc/udev/rules.d/z60_libpisock9.rules
> etc/udev/rules.d/60-persistent-input.rules
> etc/udev/rules.d/60-persistent-v4l.rules
> etc/udev/rules.d/z60_libsane-extras.rules
> etc/udev/rules.d/025_libgphoto2.rules
> etc/udev/rules.d/z60_alsa-utils.rules
> etc/udev/rules.d/80-drivers.rules
> etc/udev/pcmcia.rules
> etc/scsi_id.config
etc/modprobe.d etc/modprobe.d
etc/modprobe.d/linux-sound-base_noOSS.conf | etc/modprobe.d/oss-compat
etc/modprobe.d/aliases.conf | etc/modprobe.d/libsane
etc/modprobe.d/radeon-kms.conf | etc/modprobe.d/blacklist
etc/modprobe.d/blacklist.conf | etc/modprobe.d/libpisock9
etc/modprobe.d/libpisock9.conf | etc/modprobe.d/alsa-base
etc/modprobe.d/fbdev-blacklist.conf | etc/modprobe.d/alsa-base-blacklist
etc/modprobe.d/alsa-base.conf | etc/modprobe.d/arch
etc/modprobe.d/i915-kms.conf | etc/modprobe.d/arch/i386
etc/modprobe.d/fglrx-driver.conf | etc/modprobe.d/display_class
etc/modprobe.d/alsa-base-blacklist.conf | etc/modprobe.d/pnp-hotplug
etc/modprobe.d/oss-compat.conf | etc/modprobe.d/linux-sound-base_noOSS
etc/modprobe.d/nvidia-kernel-common.conf | etc/modprobe.d/arch-aliases
> etc/modprobe.d/aliases
lib lib
lib/modules lib/modules
lib/modules/-- lib/modules/--
lib/modules/--/modules.symbols <
lib/modules/--/modules.devname <
lib/modules/--/modules.dep.bin <
lib/modules/--/modules.dep <
lib/modules/--/modules.alias <
lib/modules/--/modules.softdep <
lib/modules/--/modules.order <
lib/modules/--/modules.alias.bin <
lib/modules/--/kernel lib/modules/--/kernel
lib/modules/--/kernel/drivers lib/modules/--/kernel/drivers
> lib/modules/--/kernel/drivers/cdrom
> lib/modules/--/kernel/drivers/cdrom/cdrom.ko
lib/modules/--/kernel/drivers/ata lib/modules/--/kernel/drivers/ata
lib/modules/--/kernel/drivers/ata/libata.ko lib/modules/--/kernel/drivers/ata/libata.ko
lib/modules/--/kernel/drivers/ata/ahci.ko lib/modules/--/kernel/drivers/ata/ahci.ko
lib/modules/--/kernel/drivers/virtio <
lib/modules/--/kernel/drivers/virtio/virtio_pci.ko <
lib/modules/--/kernel/drivers/scsi lib/modules/--/kernel/drivers/scsi
lib/modules/--/kernel/drivers/scsi/scsi_mod.ko lib/modules/--/kernel/drivers/scsi/scsi_mod.ko
lib/modules/--/kernel/drivers/scsi/sd_mod.ko lib/modules/--/kernel/drivers/scsi/sd_mod.ko
lib/modules/--/kernel/drivers/acpi lib/modules/--/kernel/drivers/acpi
lib/modules/--/kernel/drivers/acpi/fan.ko lib/modules/--/kernel/drivers/acpi/fan.ko
> lib/modules/--/kernel/drivers/acpi/processor.ko
lib/modules/--/kernel/drivers/acpi/thermal.ko lib/modules/--/kernel/drivers/acpi/thermal.ko
lib/modules/--/kernel/drivers/mmc | lib/modules/--/kernel/drivers/acpi/dock.ko
lib/modules/--/kernel/drivers/mmc/core <
lib/modules/--/kernel/drivers/mmc/core/mmc_core.ko <
lib/modules/--/kernel/drivers/mmc/card <
lib/modules/--/kernel/drivers/mmc/card/mmc_block.ko <
lib/modules/--/kernel/drivers/thermal lib/modules/--/kernel/drivers/thermal
lib/modules/--/kernel/drivers/thermal/thermal_sys.ko lib/modules/--/kernel/drivers/thermal/thermal_sys.ko
lib/modules/--/kernel/drivers/md | lib/modules/--/kernel/drivers/video
lib/modules/--/kernel/drivers/md/dm-mod.ko | lib/modules/--/kernel/drivers/video/vga16fb.ko
lib/modules/--/kernel/lib | lib/modules/--/kernel/drivers/video/vgastate.ko
lib/modules/--/kernel/lib/crc-t10dif.ko | lib/modules/--/kernel/drivers/ide
> lib/modules/--/kernel/drivers/ide/ide-cd_mod.ko
> lib/modules/--/kernel/drivers/ide/ide-core.ko
> lib/modules/--/kernel/drivers/ide/ide-disk.ko
lib/modules/--/kernel/fs lib/modules/--/kernel/fs
lib/modules/--/kernel/fs/jbd lib/modules/--/kernel/fs/jbd
lib/modules/--/kernel/fs/jbd/jbd.ko lib/modules/--/kernel/fs/jbd/jbd.ko
lib/modules/--/kernel/fs/mbcache.ko lib/modules/--/kernel/fs/mbcache.ko
lib/modules/--/kernel/fs/ext3 lib/modules/--/kernel/fs/ext3
lib/modules/--/kernel/fs/ext3/ext3.ko lib/modules/--/kernel/fs/ext3/ext3.ko
lib/modules/--/modules.symbols.bin <
lib/libselinux.so.1 lib/libselinux.so.1
lib/libblkid.so.1 <
lib/udev lib/udev
lib/udev/path_id lib/udev/path_id
lib/udev/edd_id lib/udev/edd_id
lib/udev/firmware.agent | lib/udev/cdrom_id
lib/udev/usb_id lib/udev/usb_id
lib/udev/hotplug.functions lib/udev/hotplug.functions
lib/udev/scsi_id lib/udev/scsi_id
lib/udev/rules.d | lib/udev/vol_id
lib/udev/rules.d/91-permissions.rules <
lib/udev/rules.d/50-udev-default.rules <
lib/udev/rules.d/55-dm.rules <
lib/udev/rules.d/60-persistent-storage.rules <
lib/udev/rules.d/60-persistent-storage-dm.rules <
lib/udev/rules.d/80-drivers.rules <
lib/udev/ata_id lib/udev/ata_id
lib/ld-linux.so.2 lib/ld-linux.so.2
> lib/libvolume_id.so.0
lib/libdl.so.2 lib/libdl.so.2
lib/libuuid.so.1 | lib/klibc--IOwh0VR87LX1LY95rmnFLc1vuY.so
lib/libudev.so.0 <
lib/libdevmapper.so.1.02.1 <
lib/libm.so.6 lib/libm.so.6
lib/klibc-wbwbg7GBhllIGBCTEv0okjVELWk.so <
lib/libc.so.6 lib/libc.so.6
scripts scripts
scripts/init-top scripts/init-top
scripts/init-top/udev <
scripts/init-top/all_generic_ide scripts/init-top/all_generic_ide
scripts/init-top/blacklist <
scripts/init-top/keymap scripts/init-top/keymap
scripts/init-top/ORDER | scripts/init-top/framebuffer
scripts/local-premount scripts/local-premount
scripts/local-premount/resume scripts/local-premount/resume
scripts/local-premount/ORDER | scripts/init-premount
> scripts/init-premount/udev
> scripts/init-premount/blacklist
> scripts/init-premount/thermal
scripts/functions scripts/functions
scripts/init-bottom scripts/init-bottom
scripts/init-bottom/udev scripts/init-bottom/udev
scripts/init-bottom/ORDER | scripts/local-top
> scripts/local-top/lvm
scripts/nfs scripts/nfs
scripts/local scripts/local
bin bin
bin/losetup <
bin/mount bin/mount
bin/insmod bin/insmod
bin/umount bin/umount
bin/resume bin/resume
bin/gunzip bin/gunzip
bin/cat bin/cat
bin/gzip <
bin/mkfifo bin/mkfifo
bin/ls <
bin/dd bin/dd
bin/cpio bin/cpio
bin/minips bin/minips
bin/sync bin/sync
bin/nfsmount bin/nfsmount
bin/uname bin/uname
bin/reboot bin/reboot
bin/sh bin/sh
bin/pivot_root bin/pivot_root
bin/sleep bin/sleep
bin/mkdir bin/mkdir
bin/poweroff bin/poweroff
bin/dmesg bin/dmesg
bin/nuke bin/nuke
bin/busybox bin/busybox
bin/kill bin/kill
> bin/zcat
bin/false bin/false
bin/fstype bin/fstype
bin/mknod bin/mknod
bin/ln bin/ln
bin/run-init bin/run-init
bin/chroot bin/chroot
bin/readlink bin/readlink
bin/ipconfig bin/ipconfig
bin/halt bin/halt
bin/sh.shared bin/sh.shared
bin/true bin/true
conf conf
conf/modules conf/modules
conf/arch.conf conf/arch.conf
conf/conf.d conf/conf.d
conf/conf.d/resume conf/conf.d/resume
conf/conf.d/driver-policy conf/conf.d/driver-policy
conf/initramfs.conf conf/initramfs.conf
sbin sbin
> sbin/udevsettle
sbin/modprobe sbin/modprobe
sbin/rmmod sbin/rmmod
sbin/udevadm sbin/udevadm
sbin/blkid <
sbin/udevd sbin/udevd
sbin/dmsetup | sbin/depmod
9388 blocks | 8671 blocks
<
 
Old 02-14-2011, 03:47 AM   #4
business_kid
LQ Guru
 
Registered: Jan 2006
Location: Ireland
Distribution: Slackware & Android
Posts: 7,565

Rep: Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696
So the initrd is the problem. You need only the modules to get you going - filesystem support & basic stuff

Never mind the one that's there. man mkinitrd
 
Old 02-19-2011, 03:40 PM   #5
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
I think the kernel is to blame, not the initrd image. A bug exists that does not initialize the Intel driver properly for the network card I think. And how am I ever going to fix this since I cannot upgrade a kernel under which I have no access to the repositories? A manual kernel build I guess?

I ran mkinitrd and ensured that most modules were included in the image, not just those guessed at by the utility as being needed to boot up. The image file swelled to 6 megs from about 2 megs or less, but still no luck.

I further believe the last upgrade was the culprit; it included an upgrade to Grub2 which archived my old boot menu list, which in turn did not even have a 486 kernel as an option (only a 686 kernel) and all worked ok before the upgrade. Now I have a new kernel and a new bootloader, and nothing works under the 686 kernel for networking. If it weren't for the 486 kernel I'd have a near-brick for a machine.

I forgot to include relevant dmesg output last time I wrote for the 486 and 686 kernels--certain messages are conspicuously absent from the 686 boot messages, below:
======486=====
[ 0.634259] Initalizing network drop monitor service
[ 5.626041] e1000e: Intel(R) PRO/1000 Network Driver - 1.0.2-k2
[ 5.626109] e1000e: Copyright (c) 1999-2008 Intel Corporation.
[ 5.626241] e1000e 0000:02:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[ 5.626349] e1000e 0000:02:00.0: setting latency timer to 64
[ 5.626551] e1000e 0000:02:00.0: irq 29 for MSI/MSI-X
[ 5.749216] 0000:02:00.0: eth0: (PCI Express:2.5GB/s:Width x1) 00:16:d3:23:67:f0
[ 5.749300] 0000:02:00.0: eth0: Intel(R) PRO/1000 Network Connection
[ 5.749443] 0000:02:00.0: eth0: MAC: 2, PHY: 2, PBA No: 005302-003
[ 20.016812] ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 21.574596] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX/TX
[ 21.577152] 0000:02:00.0: eth0: 10/100 speed: disabling TSO
[ 21.580157] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 32.536022] eth0: no IPv6 routers present
======686======
[ 1.546097] e1000e: Intel(R) PRO/1000 Network Driver - 0.3.3.3-k2
[ 1.549775] e1000e: Copyright (c) 1999-2008 Intel Corporation.
[ 1.549775] e1000e 0000:02:00.0: Disabling L1 ASPM
[ 1.674669] eth0: (PCI Express:2.5GB/s:Width x1) 00:16:d3:23:67:f0
[ 1.674669] eth0: Intel(R) PRO/1000 Network Connection
[ 1.674669] eth0: MAC: 2, PHY: 2, PBA No: 005302-003


They are similar but the 686 never shows the eth0 as UP, whereas the 486 does. Why? I have no idea. The versions are different, and I attacked that: I installed the kernel development package, downloaded the kernel source for the 686 with header files and then proceeded to download the Intel source code to make the latest module for this hardware. It complained of some references...it guessed wrong the location of the source code but only by a little...I fixed that...it compiled and installed cleanly. I even rmmod'ed the old module and loaded the new one. Still no luck but I did get odd behavior at that point.

If I used ifconfig to UP the eth0 and set the IP address to the address this machine is normally given, then I could ping the machine from another machine on the same network. [I don't know what algorithm is used by DHCP in the router modem I have but as long as the 2 or 3 machines I have are attached by the same ports they seem to consistently be assigned the same ip addresses every day--and under the 486 kernel, yes, the wired network is setup for DHCP and it works]. However, aside from being able to ping the machine I still have no connectivity; the gui shows none--although I've learned not to trust gui that much--and browsers cannot get anywhere, logging out and logging back in doesn't help....Rebooting didn't help. Inserting e1000e as a module to load at boot time did not help, although I have to try that one again to make sure I'm specifying the correct one since an older version is possibly lying around someplace, if I'm allowed to specify a file path that is...

Am I at a loss....this has been several days now...and 2 solid days most recently booting, rebooting, chasing leads...getting nowhere. The bug I thought might be applicable here is:
http://old.nabble.com/Bug-497392%3A-...d27686719.html
although the hardware is slightly different, but otherwise everything else is on point.

I will re-make the initrd image one last time under the 686 kernel and now that I have the Intel module compiled from Intel source code, merely to ensure it wasn't a question of getting the latest version.

Assuming it's the kernel, I've not had any luck finding older stable debian kernel versions in the 686 line? Where do I look? I suppose I could figure out how to compile and make the boot image vmlinuz someday but I cannot even start down that road without the source code.
 
Old 02-19-2011, 05:25 PM   #6
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
I made some progress, but still need a little help to get the extra point!
======PARTIAL SUCCESS=====
helix@x60:/lib/modules/2.6.26-2-686/kernel/drivers/net/e1000e$ sudo ifup eth0
Ignoring unknown interface eth0=eth0.
helix@x60:/lib/modules/2.6.26-2-686/kernel/drivers/net/e1000e$ sudo ifdown eth0
ifdown: interface eth0 not configured
helix@x60:/lib/modules/2.6.26-2-686/kernel/drivers/net/e1000e$ sudo emacs /etc/network/interfaces
[I merely added 'auto eth0' and un-commented 'iface eth0 inet dhcp' which had been prepended with a weird comment that read "#Network Manager#--so I just hit enter and made a newline after the second # sign--could that be a bug in creation script that left it all on one line?!]
helix@x60:/lib/modules/2.6.26-2-686/kernel/drivers/net/e1000e$ sudo /etc/init.d/network-manager stop && sudo /etc/init.d/networking restart && sudo /etc/init.d/network-manager start
Stopping network connection manager: NetworkManager.
Running /etc/init.d/networking restart is deprecated because it may not enable again some interfaces ... (warning).
Reconfiguring network interfaces...Internet Systems Consortium DHCP Client 4.1.1-P1
Copyright 2004-2010 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Listening on LPF/eth0/00:16:d3:23:67:f0
Sending on LPF/eth0/00:16:d3:23:67:f0
Sending on Socket/fallback
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval 5
DHCPOFFER from 192.168.1.1
DHCPREQUEST on eth0 to 255.255.255.255 port 67
DHCPACK from 192.168.1.1
bound to 192.168.1.27 -- renewal in 39873 seconds.
done.
Starting network connection manager: NetworkManager.
helix@x60:/lib/modules/2.6.26-2-686/kernel/drivers/net/e1000e$ sudo ifconfig
eth0 Link encap:Ethernet HWaddr 00:16:d3:23:67:f0
inet addr:192.168.1.27 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::216:d3ff:fe23:67f0/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:39 errors:0 dropped:0 overruns:0 frame:0
TX packets:49 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:3044 (2.9 KiB) TX bytes:7707 (7.5 KiB)
Interrupt:16 Memory:ee000000-ee020000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:70 errors:0 dropped:0 overruns:0 frame:0
TX packets:70 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:5312 (5.1 KiB) TX bytes:5312 (5.1 KiB)

The above shows real progress; the output of ifconfig is new and so is the output of the restart of the network manager--first time I've done that too. I just tested a browser under the 686 kernel and I'm on the internet!!! Only gotcha that might bite me hard later with other software is that the GUI--gnome--still reflects no connections and that I am supposedly 'disconnected'. If I can fix that and ensure that the changes I just made persist, then I'm in business.
 
Old 02-22-2011, 06:09 PM   #7
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
I have not solved this problem; the wireless appears to work under the 686 kernel only if I am plugged into a network cable at boot, otherwise it won't work at all. There are still other problems; it's not actually working correctly and continues to indicate it is not working even when it sometimes is working.

I think part of the problem is that I have a Lenny 686 kernel (2.6.26) and a Squeeze 486 kernel (2.6.32). This goes a long way to account for the different behaviors of network manager, the different wireless drivers, etc.

I thought of trying this possible solution involving the removal of network manager found at this link using wicd:
http://greenbeedigital.com.au/conten...6-debian-lenny

But I'm hesitant; I'm uncertain I can remove network manager just for the 686 kernel but not for the 486 kernel, or if so how. I don't want to break what I have working under the 486 kernel else I have no connectivity at all and I might as well abandon Debian all-together.
 
Old 02-22-2011, 10:25 PM   #8
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
debug info for network manager--anyone?!

I turned on some debug level logging for the networkmanager in case that sheds light on this....anyone?!

At this moment I am using the 68 kernel but network-manager shows no network connections, despite ifconfig showing an ip address---no luck anymore on wireless at all.

lsmod only shows iw3945 loaded but modprobe shows iw4965 as well; removing a reloading or just removing 4965 does not help:
kernel/drivers/net/wireless/iwlwifi/iwl4965.ko
kernel/drivers/net/wireless/iwlwifi/iwlcore.ko
kernel/drivers/net/wireless/iwlwifi/iwl3945.ko
kernel/drivers/infiniband/hw/cxgb3/iw_cxgb3.ko
kernel/drivers/infiniband/hw/amso1100/iw_c2.ko
kernel/drivers/infiniband/hw/nes/iw_nes.ko
kernel/drivers/infiniband/core/iw_cm.ko

=====686 kernel networkmanager debug logging====
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> NetworkManager (version 0.8.1) is starting...
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> Read config file /etc/NetworkManager/NetworkManager.conf
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> modem-manager is now available
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: init!
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: update_system_hostname
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPluginIfupdown: management mode: managed
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/eth0, iface: eth0)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/eth0, iface: eth0): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/irda0, iface: irda0)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/irda0, iface: irda0): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/lo, iface: lo)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/lo, iface: lo): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/pan0, iface: pan0)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/pan0, iface: pan0): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/wlan0, iface: wlan0)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/wlan0, iface: wlan0): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: devices added (path: /sys/class/net/wmaster0, iface: wmaster0)
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: device added (path: /sys/class/net/wmaster0, iface: wmaster0): no ifupdown configuration found.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: end _init.
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> Loaded plugin ifupdown: (C) 2008 Canonical Ltd. To report bugs please use the NetworkManager mailing list.
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> Loaded plugin keyfile: (c) 2007 - 2008 Red Hat, Inc. To report bugs please use the NetworkManager mailing list.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: (141579040) ... get_connections.
Feb 22 22:03:44 x60 NetworkManager[10455]: SCPlugin-Ifupdown: (141579040) connections count: 0
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> WiFi enabled by radio killswitch; enabled by state file
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> WWAN enabled by radio killswitch; enabled by state file
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> WiMAX enabled by radio killswitch; enabled by state file
Feb 22 22:03:44 x60 NetworkManager[10455]: <info> Networking is enabled by state file
Feb 22 22:03:44 x60 NetworkManager[10455]: <warn> /sys/class/net/eth0: couldn't determine device driver; ignoring...
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.279507] [nm-udev-manager.c:412] net_add(): ignoring interface with type 783
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.279570] [nm-udev-manager.c:412] net_add(): ignoring interface with type 772
Feb 22 22:03:44 x60 NetworkManager[10455]: <warn> /sys/class/net/pan0: couldn't determine device driver; ignoring...
Feb 22 22:03:44 x60 NetworkManager[10455]: <warn> /sys/class/net/wlan0: couldn't determine device driver; ignoring...
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.279937] [nm-udev-manager.c:412] net_add(): ignoring interface with type 801
Feb 22 22:03:44 x60 NetworkManager[10455]: <warn> bluez error getting default adapter: No such adapter
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.292872] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 1 flags 0x10049
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.292970] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 2 flags 0x11043
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.293008] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 3 flags 0x80
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.293043] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 4 flags 0x11043
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.293079] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 5 flags 0x1003
Feb 22 22:03:44 x60 NetworkManager[10455]: <debug> [1298430224.293122] [nm-netlink-monitor.c:117] link_msg_handler(): netlink link message: iface idx 6 flags 0x1002
 
Old 02-23-2011, 03:37 AM   #9
business_kid
LQ Guru
 
Registered: Jan 2006
Location: Ireland
Distribution: Slackware & Android
Posts: 7,565

Rep: Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696
Quote:
I have not solved this problem; the wireless appears to work under the 686 kernel only if I am plugged into a network cable at boot, otherwise it won't work at all. There are still other problems; it's not actually working correctly and continues to indicate it is not working even when it sometimes is working.
Taking this as still true, it clears your startup files, and hardware. You are trying for wifi. Have you firmware? in /lib/firmware, there is a pile of firmware associated with kernel sources these. Don't post long lists; it forces us to do needle in a haystack stuff. With no internet cable plugged in at boot,
run 'ifconfig wlan0 up' & post the output of
cat /proc/interrupts |grep -e 'wlan0' -e 'eth0'
ifconfig
iwconfig wlan0
iwlist wlan0 scan - do you see APs?
 
Old 02-23-2011, 05:10 PM   #10
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
Here's the output you requested to help diagnose this problem--this is right after boot with no cable attached:
helix@x60:~$ sudo ifconfig wlan0
wlan0 Link encap:Ethernet HWaddr 00:13:02:39:f6:ea
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

helix@x60:~$ sudo ifconfig
eth0 Link encap:Ethernet HWaddr 00:16:d3:23:67:f0
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:16 Memory:ee000000-ee020000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:70 errors:0 dropped:0 overruns:0 frame:0
TX packets:70 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:5312 (5.1 KiB) TX bytes:5312 (5.1 KiB)

helix@x60:~$ sudo ifconfig wlan0 up
helix@x60:~$ cat /proc/interrupts | grep -e 'wlan0' -e 'eth0'
219: 126 0 PCI-MSI-edge eth0

==[by manual inspection I found another line that might be relevant:
217: 29565 0 PCI-MSI-edge iwl3945
iwl3945 being the wireless driver, of course]==

helix@x60:~$ sudo ifconfig
eth0 Link encap:Ethernet HWaddr 00:16:d3:23:67:f0
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:16 Memory:ee000000-ee020000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:70 errors:0 dropped:0 overruns:0 frame:0
TX packets:70 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:5312 (5.1 KiB) TX bytes:5312 (5.1 KiB)

wlan0 Link encap:Ethernet HWaddr 00:13:02:39:f6:ea
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

wmaster0 Link encap:UNSPEC HWaddr 00-13-02-39-F6-EA-6C-6F-00-00-00-00-00-00-00-00
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)

helix@x60:~$ sudo iwconfig wlan0
wlan0 IEEE 802.11 ESSID:""
Mode:Managed Frequency:2.412 GHz Access Point: Not-Associated
Tx-Power=15 dBm
Retry min limit:7 RTS thr:off Fragment thr=2352 B
Encryption key:off
Link Quality:0 Signal level:0 Noise level:0
Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0
Tx excessive retries:0 Invalid misc:0 Missed beacon:0

helix@x60:~$ sudo iwlist wlan0 scan | more
wlan0 Scan completed :
Cell 01 - Address: 30:37:A6:C3:88:01
...
[at this point I can see access points--I've been here before--but I cannot associate to any; I tried using he iwconfig ap command with a mac address and I tried assigning an ESSID to match one of he points as well, but nothing worked, but I also got no error messages anywhere that I can recall or now find...it just does nothing.]

Plugging the cable back after this exercise while turned on did not yield a network connection, nor does ifconfig down and then up upon eth0 fix this either; I have to restart network manager for it to try to get a an IP address through DHCP as per /etc/network/interfaces in which I had to add a line for iface eth0 inet dhcp, without which I'd have zero connectivity all the time. I already had auto eth0 and before that merely allow-hotplug eth0, but again, I get no connectivity with just those lines in the interfaces file.

Thanks so much.
 
Old 02-24-2011, 03:36 AM   #11
business_kid
LQ Guru
 
Registered: Jan 2006
Location: Ireland
Distribution: Slackware & Android
Posts: 7,565

Rep: Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696
if you want to restore a wired network connection after unplugging the cable, use something like
/etc/rc.d/init.d/<rc_network_script> restart (Varying from system to system)

You might well just have an encryption problem. Are we aiming at an encrypted or open network?
If open, it's a firmware problem.
 
Old 02-24-2011, 06:58 PM   #12
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
It's a firmware issue then as I have tried to associate with an open (non-encrypted) access point. So what does this mean? I have to wait for a future next kernel release?

Some quick searching lead me to a few posts about non-free firmware being removed from the linux kernel going forward that has affected some users...possibly I can find the firmware I need and get it? And firmware is distinct from the drivers that are either built into the kernel or that I downloaded and compiled from intel?
 
Old 02-24-2011, 07:01 PM   #13
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
I forgot to ask, you suggested I "use something like /etc/rc.d/init.d/<rc_network_script> restart (Varying from system to system)"; is that "the way to do it" or more of a hack? I only ask because under the 486 kernel I have no such need for any scripts, it just works whenever I plug and unplug it. And I have to now research how to write such a script that is appropriate for this platform or where to get one.
 
Old 02-25-2011, 03:29 AM   #14
business_kid
LQ Guru
 
Registered: Jan 2006
Location: Ireland
Distribution: Slackware & Android
Posts: 7,565

Rep: Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696Reputation: 696
The scripts that set up the system are in /etc/rc.d. Your 486 kernel works because they boot the network. I'm simply sendiong a restart to your network script, which is fair enough and not really a hack. If it was, I'd be the first to admit it
 
Old 02-26-2011, 03:09 PM   #15
irish_confetti
LQ Newbie
 
Registered: Jun 2009
Posts: 15

Original Poster
Rep: Reputation: 0
I think I finally found the answer; this is a kernel bug / firmware issue going back to 2008 that has never been fixed and I might be SOL:

http://labs.kernelconcepts.de/Public...-ThinkPad-X60/

I quote the relevant passage:
Kernel 2.6.24
Since 2.6.24 has been released I tried it with some interesting results...
Also the now in vanilla kernel included iwl3945 driver is no good. It has a hard time to associate with APs, if it does it at all, and if it does in 90% of the cases I got not a single packet through. The iwl3945 drivers in 2.6.24 are increadibly old: 1.1.17 whereas the latest version is 1.2.23! But 1.2.23 does not compile with 2.6.24! Geez... what do those guys do?
Well, anyway, there is a compat-wireless project to enable this. But even with the latest version association and transfers are almost impossible - sigh.
IPW3945 is offcially not supported anymore and has the drawback of a binary only userspace daemon, but it works reliably! But IPW3945 does not compile with 2.6.24 anymore But there is help. Using the IPW3945 patch makes the driver compile and work again on 2.6.24, cool!
...
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
Kernel Source Location - pvrusb2 on Debian 5 Lenny 2.6.26-2-686 hellbillyJoker Linux - Software 1 07-09-2010 10:26 AM
Debian Kernel 2.6 686 SMP alorents Linux - Software 4 02-17-2006 08:26 PM
Kernel Panic Debian v3.1(2.6.8-2-686) MKSrivastava Debian 4 12-02-2005 09:07 AM
Cisco 350 Series on Debian kernel 2.6.8-2-686 i0l0l0s Linux - Wireless Networking 5 03-24-2005 06:49 PM
Debian kernel 2.6.7-1-686 - Error MAKE XCONFIG - Missing QTDIR BaptismOfFire Linux - Software 4 09-02-2004 07:15 PM


All times are GMT -5. The time now is 03:58 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration