Re: Netgear MA101 Rev. A dies after minutes.
Joerg Albert <[email protected]>
| Newsgroups | gmane.linux.drivers.at76c503a.user |
|---|---|
| Message-ID | <[email protected]> |
Hi, On Wed, 10 Mar 2004, Chaffinch wrote: > log below, I ran ifconfig wlan0 down/up a couple of times after the link > died. I hope you can make more sense of this than me. > > Really appreciate the help with this, thanks. > > ... > Mar 10 21:30:41 flanders at76c503.c: wlan0: state 6 -> 7 (CONNECTED) > Mar 10 21:30:51 flanders wlan0: no IPv6 routers present > Mar 10 21:31:41 flanders at76c503.c: wlan0: timeout, state 7 > Mar 10 21:31:42 flanders at76c503.c: wlan0: lost beacon bssid > 02:00:d1:ce:3a:0f > Mar 10 21:32:33 flanders at76c503.c: wlan0: state 7 -> 0 (INIT) > ... > Mar 10 21:32:33 flanders at76c503.c: wlan0: state 6 -> 7 (CONNECTED) > Mar 10 21:32:43 flanders wlan0: no IPv6 routers present > Mar 10 21:34:37 flanders at76c503.c: wlan0: timeout, state 7 > Mar 10 21:34:37 flanders at76c503.c: wlan0: lost beacon bssid > 02:00:d1:ce:3a:0f > Mar 10 21:34:57 flanders at76c503.c: wlan0: state 7 -> 0 (INIT) > ... I guess the traffic ceased before the last line in both cases. Unfortunately the log doesn't help much, the driver simply doesn't see the beacon of the peer for 10 seconds. > Mar 10 21:35:24 flanders at76c503.c: wlan0 BSS table (curr=00000000, > new=00000000): > Mar 10 21:35:24 flanders at76c503.c: wlan0 BSS table (curr=00000000, > new=00000000): > Mar 10 21:35:24 flanders at76c503.c: wlan0: state 1 -> 6 (JOINING) > Mar 10 21:35:24 flanders at76c503.c: wlan0 find_matching_bss: returned > 00000000 > Mar 10 21:35:24 flanders at76c503.c: wlan0: state 6 -> 8 (STARTIBSS) > Mar 10 21:35:24 flanders at76c503.c: using BSSID 02:00:22:5c:95:35 Strange, the peer seems to have disappeared here. No idea how to solve this problem. Olaf Glage once (April 03) had a similar one with a Toshiba NB where traffic ceased suddenly, but he needed to re-plug the USB adapter. He used a different USB adapter (503a,RFMD), but had the same USB controller revision (which may explain the difference): > > 00:0b.0 USB Controller: NEC Corporation USB (rev 01) (prog-if 10 [OHCI]) > > Subsystem: Toshiba America Info Systems: Unknown device 0001 > > Flags: bus master, medium devsel, latency 64, IRQ 11 > > Memory at f7fff000 (32-bit, non-prefetchable) [size=4K] saying that this is a really old revision , the current revison is >= 42. His solution was to use a USB-2.0 Cardbus USB host controller instead of the internal hc. /Jörg