On Thu, Jul 4, 2013 at 1:35 AM, Ilia Mirkin <imirkin@xxxxxxxxxxxx> wrote: > Hello, > > I recently got a RTL8188CUS usb adapter (which uses the rtl8192cu > driver). I'm having trouble getting it to stay associated: > > [601369.845916] wlan0: associated > [601377.106440] rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try > to reconnect now > [601383.867706] wlan0: associated > [601391.120270] rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try > to reconnect now > [601397.910678] wlan0: associated > [601405.141764] rtlwifi:rtl_watchdog_wq_callback():<0-0> AP off, try > to reconnect now > > and so on. A different usb adapter (Ralink RT2573) has no trouble > associating to that AP (but it keeps hanging and the kernel isn't > always able to reset it properly). The network in question is using > WPA2-PSK. > > Is there some setting I could adjust so that it would wait longer > before deciding that the AP is off? Or something else I can do? > > This is on a 3.10-rc6 kernel + the __queue_work warning fix patch at > https://git.kernel.org/cgit/linux/kernel/git/linville/wireless-testing.git/patch/?id=1621c9466099faf21ad19738d54a7d1845a29b32. In addition to the messages above, I also occasionally saw sequences of 3-6 rtlwifi:addbareq_rx():<100-1> sta is NULL messages after an association timeout. Here are a few things I tried: (a) increasing the timeout to 20 seconds, from 6. Didn't help. (b) printing each beacon frame (in rtl_beacon_statistic). This showed some potentially interesting results... right after association, I see beacons from various APs (including the right one). But after about 1 second, all the other APs die out, and I only see beacons from 2 related APs (foo and foo-guest, with nearly identical bssid's), neither of which is the one I want. I think that they're stronger than the AP I'm trying to connect to. Sometimes the "right" AP's beacons come back, sometimes not. (c) Realtek's driver (rtl8188C_8192C_usb_linux_v3.4.4_4749.20121105). Eventually I got it to compile. It links just fine to the AP, no wifi-related problems. (The driver hung the system on usb removal, but that seems more related to poor coding.) So the AP isn't too far away, and it's not a situation where there's too much noise for the hardware to handle. Running a ping shows that there aren't any significant connection drops. Let me know if there are other things you think I should try. Happy to test any patches/provide info. -ilia -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
