Wireless gone after upgrade [Solved]

[vaettchen]

Just upgraded my daughter's machine from 8 om a rather recent Sabayon 8 to Sabayon 9. Everything fine except that the Wireless is not working anymore'- the card is disabled, and does not react to the hardware switch (Fn-F12) or the check box in knetworkmanager.

The machine a Dell that came with Suse Linux, uses this WiFi card:

Code: Select all

07:00.0 Network controller: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) (rev 01)

Here is the lsmod output:

Code: Select all

# lsmod
Module                  Size  Used by
ipt_REJECT              1904  1
ipt_LOG                 6056  1
xt_limit                1432  2
xt_tcpudp               2064  8
xt_addrtype             2176  4
xt_state                 976  6
ip6table_filter         1104  1
ip6_tables             13828  1 ip6table_filter
nf_nat_irc              1136  0
nf_conntrack_irc        3080  1 nf_nat_irc
nf_nat_ftp              1328  0
nf_nat                 12313  2 nf_nat_irc,nf_nat_ftp
nf_conntrack_ipv4       8730  8 nf_nat
nf_defrag_ipv4          1150  1 nf_conntrack_ipv4
nf_conntrack_ftp        4984  1 nf_nat_ftp
nf_conntrack           48165  7 xt_state,nf_nat_irc,nf_conntrack_irc,nf_nat_ftp,nf_nat,nf_conntrack_ipv4,nf_conntrack_ftp
iptable_filter          1152  1
ip_tables              12773  1 iptable_filter
x_tables               13341  10 ipt_REJECT,ipt_LOG,xt_limit,xt_tcpudp,xt_addrtype,xt_state,ip6table_filter,ip6_tables,iptable_filter,ip_tables
ipv6                  237344  36
snd_hda_codec_realtek   173840  1
snd_hda_codec_hdmi     19342  1
ath9k                  66433  0
mac80211              176242  1 ath9k
fglrx                3155436  73
ath9k_common            1809  1 ath9k
ath9k_hw              309359  2 ath9k,ath9k_common
snd_hda_intel          19000  3
snd_hda_codec          60896  3 snd_hda_codec_realtek,snd_hda_codec_hdmi,snd_hda_intel
rts_pstor             312236  0
r8169                  34408  0
snd_hwdep               4382  1 snd_hda_codec
snd_pcm                55658  3 snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec
uvcvideo               50940  0
videodev               67462  1 uvcvideo
media                   7890  2 uvcvideo,videodev
hp_wmi                  6040  0
ath                    12852  3 ath9k,ath9k_common,ath9k_hw
snd_timer              13958  1 snd_pcm
pcspkr                  1472  0
btusb                   8824  0
snd                    43877  13 snd_hda_codec_realtek,snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm,snd_timer
v4l2_compat_ioctl32     7637  1 videodev
bluetooth             112619  1 btusb
cfg80211              132007  3 ath9k,mac80211,ath
snd_page_alloc          5845  2 snd_hda_intel,snd_pcm
sp5100_tco              3680  0
i2c_piix4               7520  0
i2c_core               14248  2 videodev,i2c_piix4
k10temp                 2376  0
video                   9950  0
joydev                  8064  0
sparse_keymap           2329  1 hp_wmi
xts                     2213  0
gf128mul                5338  1 xts
sha512_generic          6657  0
sha1_generic            1787  0
aes_x86_64              7340  0
virtio_pci              5575  0
virtio_blk              5098  0
virtio_net             10688  0
virtio_console         12440  0
virtio_balloon          3388  0
virtio_ring             3742  5 virtio_pci,virtio_blk,virtio_net,virtio_console,virtio_balloon
virtio                  3213  5 virtio_pci,virtio_blk,virtio_net,virtio_console,virtio_balloon
sha256_generic          9501  0
iscsi_tcp               7724  0
tg3                   115327  0
e1000                  83663  0
fuse                   54994  1
xfs                   417549  0
nfs                   246010  0
nfs_acl                 1983  1 nfs
auth_rpcgss            35290  1 nfs
lockd                  61372  1 nfs
sunrpc                163888  4 nfs,nfs_acl,auth_rpcgss,lockd
jfs                   139607  0
reiserfs              203193  0
ext4                  228944  2
jbd2                   43238  1 ext4
ext3                  106307  0
jbd                    36315  1 ext3
ext2                   53365  1
mbcache                 4513  3 ext4,ext3,ext2
scsi_wait_scan           695  0
sl811_hcd               8915  0
ohci_hcd               17358  0
uhci_hcd               18658  0
ehci_hcd               33418  0
sx8                    10885  0
imm                     8645  0
parport                26823  1 imm
pata_pcmcia             9621  0
pcmcia                 29224  1 pata_pcmcia


I have trief to fix the situation with rfkill because the machine does not react to any of the usual keys. Here is the output of

Code: Select all

rfkill list
0: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
1: hp-wifi: Wireless LAN
        Soft blocked: yes
        Hard blocked: no
2: hp-bluetooth: Bluetooth
        Soft blocked: yes
        Hard blocked: no
3: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: yes


Any attempt to rfkill unblock 1, 3 or all does not change the situation.

What can I do? She is close to hitting me on the head...

Thanks,
Rainer

[vaettchen]

Some more facts:

1. I tried with the installation disk that I had originally used, i.e. wireless worked out of the box at the time. Now, however, also this old live version cannot access the WLAN card. I tried with some other Debian based Live disk - same result. So something serious has happened to the access to the wireless card that cannot be attributed to the latest version of Sabayon - which is good news.

2. But since this blackout happened during the upgrade process (i.e. before "equo upgrade", the WLAN was working perfectly, and afterwards gone), there appears to be a relation - something during the upgrade process seems to have caused the cord to be cut between the WLAN card and the rest of the laptop (HP 635, by the way).

3. It definitely seems to be a rfkill issue, as the two output shows:

Code: Select all

# iwconfig wlan0
wlan0     IEEE 802.11bgn  ESSID:off/any 
          Mode:Managed  Access Point: Not-Associated   Tx-Power=off   
          Retry  long limit:7   RTS thr:off   Fragment thr:off
          Encryption key:off
          Power Management:off


so the card is there and can be "seen" by the system but:

Code: Select all

# ifconfig wlan0 up
SIOCSIFFLAGS: Operation not possible due to RF-kill

For what it's worth, some snippets from dmesg:

Code: Select all

# dmesg | grep ath
[   11.645425] ath9k 0000:07:00.0: PCI INT A -> GSI 23 (level, low) -> IRQ 23
[   11.645443] ath9k 0000:07:00.0: setting latency timer to 64
[   11.696013] ath: EEPROM regdomain: 0x60
[   11.696019] ath: EEPROM indicates we should expect a direct regpair map
[   11.696026] ath: Country alpha2 being used: 00
[   11.696028] ath: Regpair used: 0x60
[   11.698800] ieee80211 phy0: Selected rate control algorithm 'ath9k_rate_control'
[   11.699125] Registered led device: ath9k-phy0
[   12.115344] usbcore: registered new interface driver ath3k

and selected output from /var/log/messages:

Code: Select all

Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): using nl80211 for WiFi device control
Jun  8 06:55:03 localhost NetworkManager[8106]: <warn> (wlan0): driver supports Access Point (AP) mode
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): new 802.11 WiFi device (driver: 'ath9k' ifindex: 3)
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): exported as /org/freedesktop/NetworkManager/Devices/1
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): now managed
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): device state change: unmanaged -> unavailable (reason 'managed') [10 20 2]
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): bringing up device.
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> (wlan0): deactivating device (reason 'managed') [2]
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> NetworkManager is running with OpenRC...
Jun  8 06:55:03 localhost NetworkManager[8106]: <info> modem-manager is now available
Jun  8 06:55:03 localhost NetworkManager[8106]: <warn> bluez error getting default adapter: The name org.bluez was not provided by any .service files
Jun  8 06:55:04 localhost /etc/init.d/NetworkManager[8079]: WARNING: NetworkManager has started, but is inactive
...
Jun  8 06:56:37 localhost NetworkManager[8106]: <info> WiFi hardware radio set enabled
...
Jun  8 06:56:58 localhost NetworkManager[8106]: <info> (wlan0): bringing up device.
...
Jun  8 07:16:21 localhost NetworkManager[8106]:    SCPlugin-Ifnet: update_connection_setting_from_config_block: name:Denenchofu, id:Denenchofu, uuid: 5e39a597-876a-a0a4-749e-2d1cdc8bcf90
Jun  8 07:16:21 localhost NetworkManager[8106]:    SCPlugin-Ifnet: Successfully updated Denenchofu


"Denenchofu" is the SSID, so some communication seems to take place.



Any help would be greatly appreciated!

Thanks,
Rainer

[Fitzcarraldo]

Try removing the hp_wmi module to see if that then enables you to use rfkill:

Code: Select all

# rmmod hp_wmi
# rfkill list
0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
# rfkill block 0
# rfkill list
0: phy0: Wireless LAN
        Soft blocked: yes
        Hard blocked: no
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
# rfkill unblock 0
# rfkill list
0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
# rfkill block wlan
# rfkill list
0: phy0: Wireless LAN
        Soft blocked: yes
        Hard blocked: no
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
# rfkill unblock wlan
# rfkill list
0: phy0: Wireless LAN
        Soft blocked: no
        Hard blocked: no
1: hci0: Bluetooth
        Soft blocked: no
        Hard blocked: no
#

If it is the hp_wmi module that is causing the problem, blacklist it so that it doesn't get loaded at boot.

[vaettchen]

Thanks, Fitzcarraldo - I have tried that but no success.

Out of desparation, I tried many weird things, and I believe I finally found the gotcha: Pressing the hard kill switch ( Fn + F12 ) very early in the boot process seems to toggle the kill switch in a way that it is persistant, i.e. it does not react to later attempts with Fn+F12, or with the rfkill utility.

As an academic exercise, I should run a set of trials and verify the assumption, but I'm happy enough that it works...

Thanks for trying to help,
Rainer