Atheros-Broken??
-
Hey there.
I've searched the forums and haven't really found an answer for this. I have an atheros-based card, one antenna. The problem occurs in ALL modes (AP, Client, Ad-hoc). the card seems to work with pfSense, but only for a couple of seconds/minutes, depending on the phase of the moon or something crazy like that. The weird thing, is that if I release/renew the link, the card works fine for a little while, then craps out again. I'm getting really really freakin' annoyed with it :| I'm assuming that since the card does work, even if for a few seconds through minutes, then it isn't the card. (Guessing) It is something pfSense/BSD/drivers are doing shortly after acquiring an 'up' state.I'm spoiled by windows, with very little *nix experience, and haven't quite figured out how to ssh into the router or any of that jazz, nor would I really know much of what to do once I did… but if I at least knew what was wrong, and if there is some kind of work-around for it, I can at least make an honest attempt!
Little bit of info from what little I could glean between release/renew clicks and trying to search the forum on this flaky connection...
(From logs)
Jun 15 05:06:54 routed[1573]: interface ath0 to 192.168.0.13 broken: in=18 ierr=1935 out=13 oerr=0 Jun 15 05:06:54 routed[1573]: interface ath0 to 192.168.0.13 broken: in=18 ierr=1935 out=13 oerr=0(from command prompt)
$ ifconfig -v ath0;athstats ath0: flags=8843 <up,broadcast,running,simplex,multicast>mtu 1500 inet6 fe80::211:50ff:fed4:df0a%ath0 prefixlen 64 scopeid 0x4 inet 192.168.0.13 netmask 0xffffff00 broadcast 192.168.0.255 ether 00:11:50:d4:df:0a media: IEEE 802.11 Wireless Ethernet autoselect mode 11g (OFDM/18Mbps) status: associated ssid Wireless channel 11 (2462) bssid 00:18:f3:95:e6:36 authmode OPEN privacy OFF deftxkey UNDEF powersavemode OFF powersavesleep 100 txpowmax 32 txpower 32 rtsthreshold 2346 mcastrate 1 fragthreshold 2346 pureg protmode RTSCTS wme burst roaming AUTO bintval 100 AC_BE cwmin 4 cwmax 10 aifs 3 txopLimit 0 -acm ack cwmin 4 cwmax 10 aifs 3 txopLimit 0 -acm AC_BK cwmin 4 cwmax 10 aifs 7 txopLimit 0 -acm ack cwmin 4 cwmax 10 aifs 7 txopLimit 0 -acm AC_VI cwmin 3 cwmax 4 aifs 2 txopLimit 94 -acm ack cwmin 3 cwmax 4 aifs 2 txopLimit 94 -acm AC_VO cwmin 2 cwmax 3 aifs 2 txopLimit 47 -acm ack cwmin 2 cwmax 3 aifs 2 txopLimit 47 -acm 4 beacon miss interrupts 277 tx management frames 15 tx frames discarded prior to association 58 tx stopped 'cuz no xmit buffer 56 tx failed 'cuz bogus xmit rate 3 tx frames with rts enabled 14 tx frames with 11g protection 37745 rx failed 'cuz of FIFO overrun 10 rx failed 'cuz frame too short 274 rx management frames 1810 beacons transmitted 42 periodic calibration failures Antenna profile: [2] tx 193 rx 12346 [3] tx 94 rx 1</up,broadcast,running,simplex,multicast>So… anyone have any idea how to fix this? Card works (freakin fast, too)... when it works.
Noticed the most errors are on that "37745 rx failed 'cuz of FIFO overrun" line... erm... that sounds like driver or hardware is wacked out... fifo over-run i understand, though i fail to see why it is happening....
-
oh… an addition to the last post....
i have no clue why, but it seems that if I do something to keep the data streaming across the connection, it works fine till a few seconds after it stops... i.e. I can renew the link, start a download, and i can download fine, with no drops. Just downloaded a linux iso, and it didn't crap on me at all... until it finished, then i went to hceck a web-page, and it was dead again. odd thing is it still says that the link is up and associated...
ArGH i'm confuddled.
-
What pfsense version you using?
-
You must have newer madwifi-ng drivers. I know that at least the drivers from december ( or perhaps a little earlier ) till a month ago ( or maybe even current ) have this problem. I also have an atheros card and it drops after 10 min or so everytime regardless of the mode it is in. You could try installing the latest madwifi-ng drivers ( maybe svn sources ), however I am not sure how to do that on PFSense.
-
pfsense is FreeBSD, not Linux, so madwifi isn't going to help.
They do share some code, so there could be similar issues on both. That's why I asked about the version.
-
The HAL is usually updated in freebsd before it is in madwifi, so any problems fixed in madwifi should also be fixed in freebsd.
-
Sorry about the very late reply… I've had tons of stuff going on lately and haven't even had a chance to sit down and reply to this.
I'm using release 1.01. (Waiting for the next stable release to come out!!)
-
using 1.2RC2
i am experiencing kind of a similar problem. But so far the connection alway's came back.
I'm setting up a WLAN-Bridge between two buildings. I'll keep an eye on the log and try to see if i can find out when and why it gets broken.Aug 30 23:20:39 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:20:39 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:20:19 routed[888]: interface ath0 to 10.0.4.1 broken: in=9 ierr=14 out=8 oerr=0 Aug 30 23:20:19 routed[888]: interface ath0 to 10.0.4.1 broken: in=9 ierr=14 out=8 oerr=0 Aug 30 23:19:28 dnsmasq[7068]: reading /var/dhcpd/var/db/dhcpd.leases Aug 30 23:18:45 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:18:45 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:18:36 routed[888]: interface ath0 to 10.0.4.1 broken: in=12 ierr=16 out=5 oerr=0 Aug 30 23:18:36 routed[888]: interface ath0 to 10.0.4.1 broken: in=12 ierr=16 out=5 oerr=0 Aug 30 23:17:38 check_reload_status: reloading filter Aug 30 23:17:34 php: /interfaces_opt.php: Creating rrd update script Aug 30 23:17:32 kernel: ath0: link state changed to UP Aug 30 23:17:32 kernel: ath0: link state changed to DOWN Aug 30 23:17:30 kernel: ath0: link state changed to UP Aug 30 23:17:28 check_reload_status: reloading filter Aug 30 23:17:23 php: /interfaces_opt.php: Creating rrd update script Aug 30 23:17:22 routed[888]: interface ath0 to 10.0.4.1 turned off Aug 30 23:17:22 routed[888]: interface ath0 to 10.0.4.1 turned off Aug 30 23:17:22 kernel: ath0: link state changed to DOWN Aug 30 23:12:25 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:12:25 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:12:14 kernel: ath0: link state changed to UP Aug 30 23:10:31 routed[888]: interface ath0 to 10.0.4.1 broken: in=0 ierr=139 out=0 oerr=0 Aug 30 23:10:31 routed[888]: interface ath0 to 10.0.4.1 broken: in=0 ierr=139 out=0 oerr=0 Aug 30 23:05:06 check_reload_status: reloading filter Aug 30 23:04:45 dnsmasq[7068]: reading /var/dhcpd/var/db/dhcpd.leases Aug 30 23:04:24 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:04:24 routed[888]: interface ath0 to 10.0.4.1 restored Aug 30 23:04:20 kernel: ath0: link state changed to DOWN Aug 30 23:04:18 routed[888]: interface ath0 to 10.0.4.1 broken: in=0 ierr=2 out=68 oerr=0 Aug 30 23:04:18 routed[888]: interface ath0 to 10.0.4.1 broken: in=0 ierr=2 out=68 oerr=0direct after bootup:
Aug 30 23:41:31 routed[875]: Send mcast sendto(ath0, 224.0.0.9.520): No buffer space available Aug 30 23:41:17 last message repeated 7 times Aug 30 23:41:17 miniupnpd[920]: sendto(udp_notify): No buffer space available Aug 30 23:41:01 routed[875]: Send mcast sendto(ath0, 224.0.0.9.520): No buffer space available Aug 30 23:41:01 routed[875]: Send mcast sendto(ath0, 224.0.0.9.520): No buffer space available Aug 30 23:40:47 last message repeated 7 times Aug 30 23:40:47 miniupnpd[920]: sendto(udp_notify): No buffer space available Aug 30 23:40:30 routed[875]: sendto(ath0, 224.0.0.1): No buffer space available Aug 30 23:40:30 routed[875]: sendto(ath0, 224.0.0.1): No buffer space available Aug 30 23:40:30 routed[875]: Send mcast sendto(ath0, 224.0.0.9.520): No buffer space available Aug 30 23:40:30 routed[875]: Send mcast sendto(ath0, 224.0.0.9.520): No buffer space available Aug 30 23:40:17 last message repeated 3 times Aug 30 23:40:17 miniupnpd[920]: sendto(udp_notify): No buffer space available Aug 30 23:39:57 routed[875]: interface ath0 to 10.0.4.1 restored Aug 30 23:39:57 routed[875]: interface ath0 to 10.0.4.1 restored Aug 30 23:39:52 routed[875]: interface ath0 to 10.0.4.1 broken: in=1 ierr=8 out=0 oerr=0 Aug 30 23:39:52 routed[875]: interface ath0 to 10.0.4.1 broken: in=1 ierr=8 out=0 oerr=0 Aug 30 23:38:22 sshlockout[952]: sshlockout starting up Aug 30 23:38:22 sshlockout[952]: sshlockout starting up Aug 30 23:38:22 login: login on ttyv0 as root -
Ok i've been playing around and i noticed some strange things:
The problem only occurs when the card is in client mode. Never (so far) when in AP mode.
When the problem occurs rebooting pfSense helps nothing. The problem exists right after bootup.
Rebooting the AP helps. For a while…