**** BEGIN LOGGING AT Tue Jun 19 02:59:59 2012 Jun 19 04:44:20 build #2 of sibyte is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/2 Jun 19 04:54:09 build #3 of brcm63xx is complete: Failure [failed compile_7] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/3 Jun 19 06:11:44 build #3 of at91 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/3 Jun 19 09:13:23 yey bcm63xx nand image booting Jun 19 09:52:19 juhosg * r32450 /trunk/target/linux/ramips/ (12 files in 10 dirs): Jun 19 09:52:19 ramips: rt305x: add support for the Petatel PSR-680W Wireless CDMA Router Jun 19 09:52:19 This patch adds support for the Korean made Petatel PSR-680W Wireless CDMA Router. Jun 19 09:52:19 The platform is based on Ralink RT3052. Jun 19 09:52:19 http://wiki.openwrt.org/toh/petatel/psr-680w Jun 19 09:52:19 Signed-off-by: Dmitry Shmygov Jun 19 09:52:20 juhosg * r32451 /trunk/tools/e2fsprogs/patches/002-dont-build-e4defrag.patch: Jun 19 09:52:20 tools/e2fsprogs: don't build e4defrag Jun 19 09:52:21 Compilation fails on older systems due to missing syscalls, and the Jun 19 09:52:21 e4defrag tool is not used by the build-system anyway. Jun 19 09:59:29 build #3 of lantiq is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/3 Jun 19 10:13:38 juhosg * r32452 /trunk/target/linux/ar71xx/generic/profiles/netgear.mk: Jun 19 10:13:38 ar71xx: add USB LED support by default for WNDR3700/WNDR3800/WNDRMAC Jun 19 10:13:38 This adds kmod-leds-wndr3700-usb package to a default build for Jun 19 10:13:38 WNDR3700, WNDR3800 and WNDRMAC in order to support the green USB LED. Jun 19 10:13:38 Signed-off-by: Roman A. aka BasicXP Jun 19 10:13:39 juhosg * r32453 /trunk/target/linux/ar71xx/base-files/lib/ar71xx.sh: Jun 19 10:13:40 ar71xx: correctly detect NETGEAR WNDRMAC Jun 19 10:13:40 This patch helps properly detect a WNDRMAC device. Before this Jun 19 10:13:41 patch the model is detected as "NETGEAR ?????????N". Jun 19 10:13:41 Signed-off-by: Roman A. aka BasicXP Jun 19 10:13:42 juhosg * r32454 /trunk/target/linux/ar71xx/ (9 files in 2 dirs): ar71xx: add a common function to configure USB power GPIOs Jun 19 10:13:44 juhosg * r32455 /trunk/target/linux/ar71xx/files/arch/mips/ath79/ (5 files): ar71xx: remove superfluous include of asm/mach-ath79/pci.h Jun 19 10:44:17 juhosg * r32456 /trunk/target/linux/generic/files/ (3 files in 2 dirs): generic: ar8216: allow to configure AR8327 LEDs via platform data Jun 19 10:44:18 juhosg * r32457 /trunk/target/linux/ar71xx/patches-3.3/605-MIPS-ath79-db120-fixes.patch: ar71xx: make AR8327 LEDs working on the DB120 board Jun 19 11:31:18 Could someone tell me what's up with 'enable' vs. 'enabled' in uci configs? E.g. openvpn on backfire needs 'enable' and in trunk takes both 'enable' and 'enabled'. Is the latter better/newer/more standard? Jun 19 11:32:09 yes Jun 19 11:32:19 80% use enabled Jun 19 11:33:15 I'm building fairly fancy instructions for creating a vpnserver and vpnclient for ad-hoc event networking, with shaping and all. Jun 19 11:34:25 So related question: any rough idea when Attitude Adfjustment will be the release version? I.e., is it worth wiating? Jun 19 11:34:35 june & july Jun 19 11:34:39 june / july Jun 19 11:35:10 Btw: first draft is on: http://freepad.erdgeist.org/p/eventrouter Jun 19 11:35:47 (much more text to come, but the howto actually works) Jun 19 11:35:48 what about 3.3 being EOL'd ? won't be good for atitude adjustment if it will stay on this kernel :( Jun 19 11:36:08 cool Jun 19 11:36:14 stintel: not decided yet Jun 19 11:36:51 would be nice if it would be known beforehand which kernels will get lts Jun 19 11:37:09 lts? Jun 19 11:37:16 long-term-support Jun 19 11:37:20 ah Jun 19 11:37:29 maybe they no longer call it like that Jun 19 11:38:12 stintel: they offload stabilization work to the distributors anyway Jun 19 11:38:34 so if we really want lts we should probably use a redhat kernel or sometjhing Jun 19 11:39:17 but thats just my opinion and I do not decide that Jun 19 11:41:34 stintel: you often know beforehand; usually somebody volunteered maintaining it (like ubuntu for 3.2, or greg for 3.0) Jun 19 11:43:05 jow_laptop: heh no thanks for redhat kernel ;) Jun 19 11:43:19 I prefer not to go back in time ;) Jun 19 11:43:35 although they do a quite nice job backporting all kinds of niceness Jun 19 11:44:20 KanjiMonster: ic, I did not know that Jun 19 11:47:46 however given that we plan to release every 6 months from now on I wonder why we need an lts kernel Jun 19 11:51:16 juhosg * r32458 /trunk/target/linux/ar71xx/patches-3.3/ (3 files): ar71xx: refresh patches Jun 19 11:51:18 juhosg * r32459 /trunk/target/linux/ar71xx/ (3 files in 3 dirs): ar71xx: add support for the TL-WDR4300 board Jun 19 11:51:20 juhosg * r32460 /trunk/target/linux/ar71xx/base-files/ (5 files in 4 dirs): ar71xx: add user space support for the TL-WDR4300 Jun 19 11:51:20 Are there any major changes planned to OpenWRT? Or is everything more or less staying in the same place? Jun 19 11:51:21 juhosg * r32461 /trunk/target/linux/ar71xx/ (generic/profiles/tp-link.mk image/Makefile): ar71xx: create profile and build image for the TL-WDR4300 board Jun 19 11:51:43 ropg: configuration wise not Jun 19 11:51:58 jow: I like that... Jun 19 11:53:25 wdr4300 800 mbps hardware nat..? Jun 19 12:14:55 <[florian]> dape-work: some atheros switches allow you to do hardware nat Jun 19 12:15:14 very interesting, so no more cpu involved? Jun 19 12:15:40 <[florian]> less CPU intervention required at least Jun 19 12:16:00 got it, okay, i can only wait to be available on our local market.. Jun 19 12:16:04 <[florian]> but I don't think we are supporting that yet Jun 19 12:16:19 <[florian]> not quite sure how that fits into linux's networking stack nicely Jun 19 12:17:02 its complicated then Jun 19 12:17:23 which makes me wonder what kind of firmware are they using.. linux ? Jun 19 12:18:25 <[florian]> oh, on the switch, I think it might be entirely hardware Jun 19 12:18:38 <[florian]> the rest of the system certainly runs Linux Jun 19 12:19:04 <[florian]> which makes me wonder, how that works with Linux's networking stack Jun 19 12:19:18 <[florian]> whether Linux instructs the switch with the conntrack table, or the switch learns it by itself Jun 19 12:19:47 it has to be set up by the kernel Jun 19 12:20:35 juhosg probably has serial access to wdr4300 :) Jun 19 12:21:10 <[florian]> nbd: how is that done right now? the ethernet driver is hooked into conntrack and program switch entries? Jun 19 12:21:49 in openwrt? not at all. in the reference software? through ugly hackery Jun 19 12:21:57 <[florian]> in the reference software of course Jun 19 12:30:05 build #2 of etrax is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/etrax/builds/2 Jun 19 13:26:06 the lack of random on my Atheros AR7241 based (Ubiquiti Picostation) still worries me. Even if it passes the rngtool test Jun 19 13:26:12 there is no /dev/hwrandom Jun 19 13:26:48 root@vpnclient:~# cat /proc/sys/kernel/random/entropy_avail Jun 19 13:26:48 58 Jun 19 13:27:01 That's not good Jun 19 13:27:17 there are no entropy sources Jun 19 13:27:41 Doe sthe Atheros AR7241 have hwrandom? Jun 19 13:27:53 (Hard to find specs on this ^#&^@&*^@& Atheros webiste Jun 19 13:27:57 no Jun 19 13:28:30 can we harvest from NICs ? (I know, second-best source) Jun 19 13:29:07 kernel.org deemed NICs insecure entropy sources and removed them from thep ool Jun 19 13:29:38 if you do it right, an insecure source of random can never make things worse, it can only make them the same. Jun 19 13:30:06 you do not need to discuss this with me, the decision was made by kernel.org Jun 19 13:30:19 and kernel.org has it easy, most machines have so many more and better sources than we do on embedded systems Jun 19 13:30:53 And there's no way for anyone to smuggle it back in then, I guess? Jun 19 13:31:38 the point has been brought up a few times but no decision has been made Jun 19 13:32:44 So people are left generating keys with a whopping 58 bits of random. Jun 19 13:33:27 I generate my keys elsewhere and embed them Jun 19 13:33:55 unless you refer to wpa, but there some control traffic noise is fed into the pool Jun 19 13:34:16 the ssh host key comes to mind Jun 19 13:35:09 Ah, i'll just wget http://www.random.org/cgi-bin/randbyte?nbytes=400&format=h :) Jun 19 13:46:30 the packaged rngd in rng-tools does something sketchy, I think. Jun 19 13:46:37 It does "/sbin/rngd -r /dev/urandom -W 4000 -t 30" Jun 19 13:46:59 i.e. seeding the random device from pseudorandom based on the same low number of bits Jun 19 13:47:19 which "solves" the problem of things fussing about low entropy, sure.... Jun 19 14:27:11 build #2 of adm5120 is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/2 Jun 19 14:36:58 build #3 of uml is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/3 Jun 19 14:49:03 florian * r32462 /trunk/target/linux/mcs814x/ (76 files in 27 dirs): (log message trimmed) Jun 19 14:49:03 [mcs814x] add Moschip MSC814x support Jun 19 14:49:03 This target currently only supports Moschip's MCS8140 SoC, but support Jun 19 14:49:03 for other chips in the same family (MCS8142, MCS8144) will be easy to add. Jun 19 14:49:03 Target support is entirely using Device Tree for probing peripherals. Jun 19 14:49:04 Drivers support include: Jun 19 14:49:04 - PCI Jun 19 15:25:25 build #3 of brcm47xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/3 Jun 19 16:42:50 anyone using LuCI's httpclient? it seems to ignore or not properly use the "sndtimeo" and "rcvtimeo" options. if i request something from an IP that's down, it'll sit there for minutes, blocking, trying to fetch it Jun 19 16:46:23 it seems the httpclient.lua sees the timeouts fine, but maybe there is a bug in nixio in the setsockopts? Jun 19 16:54:41 build #4 of avr32 is complete: Failure [failed compile_6] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/4 Jun 19 17:03:27 jow * r32463 /trunk/package/ppp/patches/ (28 files): [package] ppp: annotate most patches with description and S-o-b Jun 19 20:24:59 hi Jun 19 20:25:03 hi Jun 19 20:25:26 are there any plans to update the current targets to 3.4? Jun 19 20:25:47 3.3 seems to be end of life so its somehow a dead road Jun 19 22:34:51 acinonyx * r32464 /trunk/package/busybox/config/coreutils/Config.in: [package] busybox: Enable ibs, obs and conv options on 'dd' Jun 19 22:36:55 acinonyx * r32465 /trunk/target/linux/x86/base-files/lib/ (preinit/89_move_config upgrade/platform.sh): Jun 19 22:36:55 [x86] Fix sysupgrade for squashfs, jffs2 and ext4 images Jun 19 22:36:55 * use 'fsync' with 'dd' for image writing Jun 19 22:36:55 * save config to boot partition **** ENDING LOGGING AT Wed Jun 20 02:59:59 2012