**** BEGIN LOGGING AT Thu Jan 14 02:59:57 2010 Jan 14 08:37:26 xMff: wpa_supplicant.sh is broken :( https://dev.openwrt.org/browser/trunk/package/wpa_supplicant/files/wpa_supplicant.sh?rev=19119#L25 seems you wanted to replace the if from my patch with [] &&, but you missed the if. Jan 14 08:38:36 http://openwrt.pastebin.ca/1750729 Jan 14 08:38:48 but that's pretty trivial :-) Jan 14 11:02:09 good morning Jan 14 12:27:48 florian * r19125 /trunk/toolchain/uClibc/ (32 files in 3 dirs): [toolchain] add support for uClibc 0.9.30.2 Jan 14 12:32:29 jow * r19126 /packages/net/haproxy/ (Makefile files/haproxy.init): [packages] haproxy: forgot to invoke start-stop-daemon with background (-b) mode Jan 14 13:19:19 xMff: sorry i'm packed at work Jan 14 13:19:24 with work Jan 14 14:01:22 danage: metoo Jan 14 16:03:03 I'm (hopefully) finishing up my biggest project to date using OpenWRT. Thanks to everyone for your work/support! Jan 14 16:03:17 for those like me who like to see pictures: http://www.psyc.vt.edu/openwrt/server_network.jpg Jan 14 16:04:35 Its being installed this afternoon. They are proxy-arp routers setup in a failover configuration. Jan 14 16:16:00 is that x86 ? Jan 14 16:16:16 yeah, its 2x alix boards Jan 14 16:16:29 nice Jan 14 16:16:56 :) Jan 14 16:23:02 just curious, is the enclosure self-built or something readymade? We planned to put some embedded hw into a datacenter (for various tasks routing/sensoring tasks) Jan 14 16:23:26 but so far there was no suitable option to mount them in the rack Jan 14 16:24:11 I got it here: http://www.netgate.com/product_info.php?cPath=67_109&products_id=774 Jan 14 16:24:23 mine doesn't have internal power though, guess they upgraded it Jan 14 16:26:22 they used to sell just the enclosure if you already have the alix boards, but I dont see it off hand Jan 14 16:28:33 yep, the only two ready ones seem to be pre-equipped with boards already Jan 14 16:30:21 mine was a custom order, I just emailed them what I wanted and they gave me the price, you could probably do that too Jan 14 16:31:57 I see Jan 14 16:45:56 All, I have a problem with the mailing list. Jan 14 16:46:14 I've crossposted to another mailing list (lacie-nas) Jan 14 16:46:51 and the guys have answered but are blocked because they are not registered in the openwrt list Jan 14 16:47:09 so the mail thread is out Jan 14 16:49:03 I will forward their mails Jan 14 16:49:17 is it possible to open the list ? Jan 14 17:26:20 tmonjalo and make it forward even more spam? Jan 14 17:26:41 i think you just learned why crossposting is a bad idea Jan 14 17:27:01 roh: you are right Jan 14 17:27:22 roh: but I have to transmit mails I think Jan 14 17:27:38 np Jan 14 17:27:38 they are interesting Jan 14 17:29:49 sorry for the spam Jan 14 17:36:08 netprince: nice board, if it just comes with gigabit ethernets... Jan 14 17:36:45 * solca wonders if it exists a board with tons of ethernets for constructing an OpenWRT layer2 ethernet switch Jan 14 17:40:30 solca: gige would be nice, but the cpu would have to be better too Jan 14 17:47:18 no serious switch routes traffic via the cpu Jan 14 17:47:29 not even when routing on >few mbits Jan 14 17:47:43 thats what switch fabrics are for. Jan 14 17:49:18 roh: right, ebtables is very powerful but sadly it has to hit the cpu :( Jan 14 17:49:53 a driver with filtering hooks in hw to ebtables probably would solve it Jan 14 17:50:14 but that infrastructure doesn't exists and would be probably very invasive Jan 14 18:25:06 sure Jan 14 18:43:07 jow * r19127 /trunk/package/dropbear/ (Makefile files/dropbear.init): [package] dropbear: add GatewayPorts (-a) option (#6503) Jan 14 18:48:43 jow * r19128 /trunk/package/wpa_supplicant/files/wpa_supplicant.sh: [package] wpa_supplicant: fix two syntax errors in wpa_supplicant.sh (#6504) Jan 14 18:50:31 that was fast :-) Jan 14 18:51:07 jow * r19129 /packages/net/coova-chilli/Makefile: [packages] coova-chilli: update to v1.2.0 (#6501) Jan 14 18:52:11 yep Jan 14 18:52:56 evening Jan 14 18:53:01 hey Jan 14 18:53:27 last nite I suddenly noticed my wifi was gone :P Jan 14 18:55:12 jow * r19130 /packages/libs/libnetfilter-conntrack/Makefile: [packages] libnetfilter-conntrack: update to v0.0.101 (#6466) Jan 14 18:55:44 jow * r19131 /packages/net/conntrack-tools/Makefile: [packages] conntrack-tools: update to v0.9.14 (#6466) Jan 14 18:58:19 jow * r19132 /packages/libs/libevent/Makefile: [packages] libevent: update to v1.4.13 (#6468) Jan 14 18:59:41 could anyone have a look at https://dev.openwrt.org/ticket/5960 ? it contains a patch to fix ASSERTION FAILED: src != NULL at fs/mini_fo/mini_fo.h:422 when executing the "write" command in quagga. Jan 14 19:00:07 jow * r19133 /packages/utils/dash/ (. Makefile): [packages] add dash - Debian Almquist Shell (#6469) Jan 14 19:02:21 jow * r19134 /packages/net/ntpclient/ (Makefile files/ntpclient.config files/ntpclient.hotplug): [packages] ntpclient: add option to bind ntpclient to a specific interface (#6470) Jan 14 19:03:26 jow * r19135 /packages/net/proftpd/Makefile: [packages] proftpd: update to v1.3.2c (#6471) Jan 14 19:08:26 stintel: ok Jan 14 19:08:48 xMff: would be nice, I've been using the patch since I commented on that ticket. seems to work like charm Jan 14 19:14:45 stintel: do you know whether quagga(-stable) needs it as well? Jan 14 19:15:03 xMff: yes Jan 14 19:15:22 xMff: well, at the date I posted that comment, I was running stable and moved to unstable because the patch was for unstable Jan 14 19:15:29 I didn't bother trying if it also applied on stable Jan 14 19:15:41 it aplies to stable as well with some fuzz, rebased it already Jan 14 19:19:11 jow * r19136 /packages/net/quagga/ (Makefile patches/150-no-cross-fs-link.patch): [packages] quagga: fix mini_fo assertion triggered by cross-fs link() (#5960) Jan 14 19:19:25 jow * r19137 /packages/net/quagga-unstable/ (Makefile patches/150-no-cross-fs-link.patch): [packages] quagga-unstable: fix mini_fo assertion triggered by cross-fs link() (#5960) Jan 14 19:20:06 awesome, thanks Jan 14 23:11:14 wth: http://ask.slashdot.org/story/10/01/14/2053244/Powerful-Linux-ISP-Router-Distribution Jan 14 23:20:45 what's wrong with that? Jan 14 23:25:02 ping xMff Jan 14 23:25:17 pong cshore Jan 14 23:26:35 xMff: for the move of functions.sh etc to /lib I can't do it because my commit access is specifically for brcm63xx and packages...when the discussion's done do you think you could do that? Jan 14 23:27:29 ping [florian] Jan 14 23:27:39 I'll look into it soon Jan 14 23:27:53 xMff: no rush, I just thought I'd ask Jan 14 23:28:29 xMff: now that I've sorted out that I know what rights I have / don't have Jan 14 23:29:23 xMff: what do you think of my comitting the preinit stuff specifically to brcm63xx? (though I'll definitely ask Florian before doing that) Jan 14 23:30:48 xMff: I think it's not a great idea, but it might get it integrated sooner, myself Jan 14 23:33:44 cshore: let me read it first :) Jan 14 23:33:59 ok Jan 14 23:34:48 the docs for most if it is at http://cshore.is-a-geek.com/openwrt/preinit.html Jan 14 23:42:18 xMff: do you have time to do that now, or should I talk to you later? Jan 15 00:41:01 xMff: I was just surprised they hadn't heard of openwrt Jan 15 01:16:36 nbd * r19138 /trunk/package/iw/Makefile: iw: update to 0.9.19 Jan 15 01:16:42 nbd * r19139 /trunk/package/mac80211/ (20 files in 2 dirs): mac80211: update to wireless-testing+wireless-2.6 from 2010-01-13 Jan 15 01:16:45 nbd * r19140 /trunk/package/iw/patches/ (. 001-sync_nl80211_h.patch 100-coverage_class.patch): iw: add patch for setting coverage class / distance Jan 15 01:16:49 nbd * r19141 /trunk/package/mac80211/patches/ (3 files): ath9k: fix short slot vs long slot handling, implement support for setting the coverage class Jan 15 01:16:53 nbd * r19142 /trunk/package/mac80211/files/lib/wifi/mac80211.sh: mac80211: implement the distance setting Jan 15 01:21:22 nice Jan 15 01:22:36 enjoy Jan 15 01:22:37 ;) Jan 15 01:22:53 nbd: you're a hero! :-) Jan 15 01:25:28 and I am flashing my device twice a day last week or so :P Jan 15 01:26:31 it's wonderful, sysupgrade is a bliss, I'm offline for 3 minutes at most, and afterwards everything "just works(tm)" Jan 15 01:26:34 :-) Jan 15 01:58:26 nbd: seems I can no longer connect to my AP with r19142. Jan 15 02:57:39 wrt0 daemon.notice hostapd: wlan0: STA 00:1c:b3:4a:e8:8b IEEE 802.11: did not acknowledge authentication response Jan 15 01:58:50 nbd: maybe I *have* to set distance and/or coverage to something? Jan 15 02:02:19 are you using 5ghz? Jan 15 02:02:57 nope. 2.4, 11n, mixed-psk Jan 15 02:03:16 I tried setting distance to 10, 25 or 100, seems to give the same result Jan 15 02:03:28 nbd * r19143 /trunk/package/mac80211/patches/540-short_slot_handling.patch: mac80211: add a missing chunk of the short slot patch Jan 15 02:04:06 i'll take a look Jan 15 02:04:17 ah, setting coverage directly to 100 (something I tried random) does seem to fix it Jan 15 02:04:45 odd Jan 15 02:05:16 can you try to find the minimum setting that still works? Jan 15 02:05:33 and now I setdistance to 50 and it broke again Jan 15 02:05:41 ok Jan 15 02:06:48 10 works, 2 doesn't, trying few more, hold on Jan 15 02:07:12 aha, from coverage 3 it starts to work Jan 15 02:07:38 distance to my AP is <10m, 1 wall and ceiling Jan 15 02:08:14 need to know anything else ? Jan 15 02:08:26 50 still breaks it? Jan 15 02:09:13 coverage 50 works fine, but distance 50 doesn't Jan 15 02:09:32 I haven't tested them all, but it seems that coverage >=3 is ok Jan 15 02:09:51 hi Jan 15 02:10:01 maybe there's something wrong with the calculation from distance -> coverage (as I understood that's what distance does) Jan 15 02:10:37 coverage = (distance + 449) / 450; Jan 15 02:10:47 normally coverage==0 should still work Jan 15 02:11:24 nope. coverage==0 --> Jan 15 03:11:07 wrt0 daemon.notice hostapd: wlan0: STA 00:1c:b3:4a:e8:8b IEEE 802.11: did not acknowledge authentication response Jan 15 02:11:54 what kind of device/card is this? Jan 15 02:12:12 routerstation pro, ubiquiti sr71-a, phy0: Atheros AR9160 MAC/BB Rev:0 AR5133 RF Rev:b0 mem=0xb0000000, irq=48 Jan 15 02:12:24 ugh... ar9160 Jan 15 02:12:37 i suspect that one's the reason why they put such a bogus value in the initvals for the ack timeout Jan 15 02:13:52 I'll quickly run a test on my x86 test machine which has another ath9k card Jan 15 02:14:25 non-ar9160 that is Jan 15 02:15:29 but from that calculation I should set distance to at least 901 (m) before it would work, seems odd :-) Jan 15 02:16:02 my recent svn build wouldn't go past the boot stage (DMZ is always on) after a fresh re-flash, is that a known problem? Jan 15 02:16:07 on wrt54gl Jan 15 02:18:49 Plouj: can't tell, I no longer own such device to try it on, sorry Jan 15 02:19:26 stintel: heh, I guess everyone got tired of the memory limitations, eh? Jan 15 02:20:28 I did :-) they're happily running a pre 8.09.2 image at a few customers now :-) Jan 15 02:20:44 on what hardware? Jan 15 02:20:53 well the wrt54gl's ;-) Jan 15 02:21:03 oh Jan 15 02:21:22 I'm using ubiquiti routerstation pro + sr71-a now Jan 15 02:21:54 with trunk r19142 :-) Jan 15 02:22:42 heh Jan 15 02:22:47 I don't think those are sold at my local computer store Jan 15 02:22:58 I'm thinking of turning an atom eeebox into a router Jan 15 02:23:12 that, plus a switch, if course Jan 15 02:24:46 have an atom running gentoo+xbmc, have looked for atom's with dual gigE but didn't find any yet, and then I found out about the routerstation and rs pro, and that it ran openwrt .. Jan 15 02:25:12 already sold an rspro based router to a customer as well :-) Jan 15 02:25:59 lol he just wanted to have it, I warned him that it would have to run trunk (and explained what that was:P), but he just wanted it anyway Jan 15 02:43:15 nbd: same problem on my x86 test machine, cant authenticate to AP with coverage <2 Jan 15 02:43:19 phy2: Atheros AR5416 MAC/BB Rev:2 AR2122 RF Rev:81 mem=0xf97c0000, irq=17 Jan 15 02:46:55 ok Jan 15 02:47:12 are you running HT40 or HT20? Jan 15 02:47:44 HT40- Jan 15 02:47:46 http://openwrt.pastebin.ca/1751904 Jan 15 02:48:35 also, since the new iw version, mac80211.sh no longer recognizes the .11n related parameters (just noticed that on my x86 box) Jan 15 02:48:47 I'll see if I can make a patch to fix that Jan 15 02:49:35 hmm Jan 15 02:49:41 or it's just an error in iw Jan 15 02:50:09 HT capabilities: 0x104e <-> Capabilities: 0x104e Jan 15 02:52:37 fix: http://openwrt.pastebin.ca/1751915 (but again very trivial:) Jan 15 02:55:31 nbd * r19144 /trunk/package/mac80211/files/lib/wifi/mac80211.sh: mac80211: fix ht capability detection (thx, stintel) Jan 15 02:56:52 cheers Jan 15 02:57:32 now bedtime for me, if you want me to test anything else related to the coverage/distance, just ping me here I'll read it tomorrow ;-) Jan 15 02:59:16 ok Jan 15 02:59:17 will do **** ENDING LOGGING AT Fri Jan 15 02:59:57 2010