**** BEGIN LOGGING AT Fri Jan 04 02:59:58 2013 Jan 04 03:38:27 build #140 of brcm47xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/140 Jan 04 03:40:41 build #119 of sibyte is complete: Failure [failed shell_11] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/119 Jan 04 09:00:43 build #140 of atheros is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/atheros/builds/140 Jan 04 09:14:01 blogic https://dev.openwrt.org/ticket/12731#comment:1 Jan 04 09:18:02 dape: you into the business of posting me my own posts now ? Jan 04 09:18:20 ah, it isnt visible my comment yet? Jan 04 09:19:23 the thing is even if wan ignore is 1 the messages still fill the log on my tp-link wr1043nd until i set list notinterface eth0.2 Jan 04 09:19:53 well Jan 04 09:20:35 its only a log message really Jan 04 09:20:49 yep but it fills quick in a day Jan 04 09:26:16 dape: if at all we could simply drop that message from the log Jan 04 09:26:44 i just modified /etc/config/dhcp with list notinterface eth0.2 Jan 04 09:28:29 ok Jan 04 09:28:51 you could modify the script to set notinterface for the wan interface Jan 04 09:28:53 or Jan 04 09:29:04 make it apply notinterface for the ignore interfaces Jan 04 09:32:54 blogic r35004 packages/utils/mtd-utils/Makefile * utils/mtd-utils: flash_eraseall must depend on flash_erase Jan 04 09:33:05 blogic r35005 trunk/package/network/utils/iwinfo/Makefile * [libiwinfo-lua] don't depend on lua Jan 04 09:51:58 build #123 of avr32 is complete: Failure [failed compile_9] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/123 Jan 04 10:09:09 build #128 of x86 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/128 Jan 04 10:09:56 blogic r35006 packages/ net/strongswan/Makefile openwrt-ar71xx-generic-tl-wdr4300-v1-squashfs-sysupgrade.bin * [packages] update strongwan to version 5.0.1. Jan 04 10:32:03 build #135 of cobalt is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/135 Jan 04 10:39:53 Hello guys, I've almost finished package asterisk11 for OpenWrt, there's just only one thing that really bothers me - on platforms that are non-x86 asterisk binaries are quite huge (it seems like that there's whole embedded pjproject). Jan 04 10:41:06 so .. on x86 asterisk ipk package has about 1.4 MB, on ar71xx is about 7 times bigger Jan 04 11:03:03 build #107 of iop32x is complete: Failure [failed shell_11] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/107 Jan 04 11:43:56 build #135 of ppc40x is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/135 Jan 04 11:46:09 build #104 of gemini is complete: Failure [failed shell_11] Build details are at http://buildbot.openwrt.org:8010/builders/gemini/builds/104 Jan 04 12:03:21 build #137 of ramips is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/137 Jan 04 12:11:40 kaloz: ping. Jan 04 12:59:36 anyone else had an issue where the standard iptables modules wouldn't get installed into the rootFS? Jan 04 13:00:58 a bunch of standard iptables extensions is statically linked into the iptables executable Jan 04 13:01:12 so its normal do lack a bunch of lib{ipt,xt}_foo.so Jan 04 13:01:17 s/do/to/ Jan 04 13:01:48 oh Jan 04 13:01:50 crap Jan 04 13:01:55 well that explains everything Jan 04 13:02:16 I had to disable static in 1.4.17 because it was dying horribly Jan 04 13:02:31 what's the reason for compiling those libs in statically? Jan 04 13:02:40 saves a lot of space Jan 04 13:02:56 really? even inside SquashFS? Jan 04 13:02:58 and runs faster due to no dlopen overhead Jan 04 13:03:01 yes Jan 04 13:03:38 well alrighty then :) Jan 04 13:03:53 also, on that note, why aren't we doing symlinks to xtables-multi in that case (at least if also building ipv6 support) Jan 04 13:04:04 since iptables and ip6tables would basically be the same binary Jan 04 13:04:40 https://dev.openwrt.org/changeset/30675 Jan 04 13:04:44 https://dev.openwrt.org/changeset/30676 Jan 04 13:04:49 those implement the static linking Jan 04 13:18:05 jow_laptop https://dev.openwrt.org/ticket/12731#comment:2 i am using pppoe and seeing the log messages Jan 04 13:18:32 maybe you guys can uncomment by default in /etc/config/dhcp the list notinterface option with the external iface name or something Jan 04 13:25:36 dape: try this: http://pastebin.com/TaG0k2Ed Jan 04 13:26:19 dape: afterwards grep no-dhcp /var/etc/dnsmasq.conf Jan 04 13:26:20 you mean without touching /etc/config/dhcp anymore Jan 04 13:26:24 yes Jan 04 13:26:32 okay, let me pwn a wr1043nd Jan 04 13:27:44 correction: http://pastebin.com/xzHQPymd Jan 04 13:27:47 missed a return Jan 04 14:09:08 jow r35007 trunk/package/network/utils/iwinfo/ src/iwinfo_nl80211.c Makefile Jan 04 14:09:08 iwinfo: don't use the txpower value from debugfs for now, it does not match the values reported via wext Jan 04 14:24:07 jow r35008 packages/openwrt-ar71xx-generic-tl-wdr4300-v1-squashfs-sysupgrade.bin * packages: remove accidentqally committed binary Jan 04 14:26:05 jow_laptop: so, if iptables is built with --enable-static, why do the additional user-space libs get copied into the rootFS? Jan 04 14:26:33 you mean libxtables, libiptc & friends? Jan 04 14:26:42 no, I mean the extensions Jan 04 14:26:51 which ones do you refer to? Jan 04 14:27:02 actually, just ip6tables Jan 04 14:27:07 $(CP) libip6t_*.so $(1)/usr/lib/iptables/ \ Jan 04 14:27:17 that's in the install def Jan 04 14:27:22 Olipro: i don't think --enable-static means 'link stuff into executables statically' Jan 04 14:27:35 it's more like 'build static libraries' Jan 04 14:27:56 nbd: I'm just referring to the iptables extensions Jan 04 14:28:02 not libiptc or libxtables and friends Jan 04 14:28:09 right, the static linking is patches from us, unrelated to --enable-static Jan 04 14:28:43 it does look to me like --enable-static at least in 1.4.17 will link all extensions, targets/matches into the executable Jan 04 14:28:52 because it doesn't build any .so files at all Jan 04 14:29:58 given that iptables/install in the Makefile does not copy the extensions libs, I suspect it's a mistake that ip6tables/install does Jan 04 14:31:39 i did not look at 1.4.17 Jan 04 14:31:50 1.4.17 has iptables / ip6tables as multicall binary Jan 04 14:31:53 but in 1.4.10, --enable-static just means build static libraries Jan 04 14:32:05 cyrusff: I know, I make symlinks to xtables-multi Jan 04 14:32:27 * nbd smells incoming default configuration image bloat... Jan 04 14:32:39 nbd: why? Jan 04 14:32:40 ;) Jan 04 14:32:44 * jow_laptop brings up this topic since months Jan 04 14:32:58 cyrusff: if all extensions are built-in as well... Jan 04 14:33:04 ah that, lol Jan 04 14:33:23 well we should update to 1.4.17 anyway once all targets have a recent enoughj kernel Jan 04 14:33:24 i'm so looking forward to nftables Jan 04 14:33:43 heh, the jury's out on nftables Jan 04 14:33:44 the kernel/user side extensions design of netfilter is stupid Jan 04 14:33:58 * Olipro mumbles something about xtables2 Jan 04 14:34:00 wasn't nftables similar to BPF? Jan 04 14:34:07 in its design Jan 04 14:34:13 there are some similarities, yes Jan 04 14:34:43 userspace has a lot more flexibility because the kernel matching code is more simple and generic Jan 04 14:35:23 i think it was just a vm similar to BPF and you generate the actual rules into bytecode in userspace so you don't need a stupid kernel extension for everything, right? Jan 04 14:35:41 right Jan 04 14:36:12 kernel doesn't even have to know about all the header fields and stuff, userspace can encode this into generic bitfield matching ops Jan 04 14:37:29 Olipro: xtables2 just changes the low level interface between userspace and kernel for setting up rules Jan 04 14:37:53 Olipro: it does nothing to clean up the extension madness Jan 04 14:38:15 yeah, some people like their xtables though Jan 04 14:38:30 I have that feeling we'll end up with Netfilter and Netfilter-NG Jan 04 14:38:48 i think there will be a compat layer for nftables Jan 04 14:38:59 for people that want to stick with the old scripts Jan 04 14:39:40 once nftables becomes usable, i will give it a try in openwrt Jan 04 14:40:28 and i will try to fix any issues that show up so we can get rid of the old netfilter as soon as possible Jan 04 14:40:31 yeah, finally something netlink based and we can get rid of all the libraries and so on Jan 04 14:42:14 nbd: btw, are you aware that GitHub mirrors your OpenWRT um... mirror Jan 04 14:42:26 yes Jan 04 14:42:39 you should ask them to do packages.git too :) Jan 04 14:47:06 we eventually planned to move our packages feed to github anyway Jan 04 14:48:42 jogo r35009 trunk/target/linux/ (143 files in 2 dirs) * bcm63xx: remove 3.3 support Jan 04 14:48:49 jogo r35010 trunk/target/linux/brcm63xx/patches-3.6/436-MTD-bcm63xxpart-allow-passing-a-caldata-offset.patch * bcm63xx: Fix caldata partition mtd patch Jan 04 14:48:56 jogo r35011 trunk/target/linux/ (7 files in 6 dirs) * bcm63xx: Add Huawei HG556a support Jan 04 14:49:24 jow_laptop: yeah, I was hoping that'd happen - Pull requests galore Jan 04 14:49:48 well, we also plan to outsource the maintenance to volunteers Jan 04 14:50:15 just because it lies on github and contributers can easily make pull requests does not imply that its getting any easier for us Jan 04 14:51:07 maybe the actual scm gruntwork becomes easier but the management aspect is not Jan 04 14:51:43 also there's a lot of "subsystems" like the pulseaudio / ffmpeg stuff where no developer has any real testcase Jan 04 14:51:52 so it would be good if it would be maintained by its powerusers Jan 04 14:52:51 right, I figured there'd be a handoff Jan 04 15:19:01 If someone's interested - there's asterisk11 patch - http://suzelly.opf.slu.cz/~sla463/asterisk-11.1.2.patch Jan 04 15:19:17 I am testing it on brcm and x86 devices Jan 04 15:19:28 strasidlo: why not put it in a Git repo? Jan 04 15:20:06 * Olipro enjoys Git remotes so much that patch files make him sad Jan 04 15:21:08 Well, I have this package in my svn/git repo Jan 04 15:21:41 throw it on Github :) Jan 04 15:23:12 I think just for those little things one patch does not need to be versioned. Jan 04 15:25:42 it's convenient for sharing Jan 04 15:25:48 and besides, why not Jan 04 15:26:17 just because it's something small, you can always squash it into a larger commit later Jan 04 15:26:58 because I have my own repo and for those, who do not want to use my whole repo it's easier to use Jan 04 15:27:28 just patch -p0 cyrus r35012 trunk/package/network/ config/firewall/files/firewall.config config/firewall/Makefile Jan 04 16:00:28 firewall: Add ULA site border for IPv6 traffic Jan 04 16:00:28 This prevents private traffic from leaking out to the internet Jan 04 16:11:17 nbd r35013 trunk/package/uci/Makefile * uci: update to latest version, adds list delete, removes unused plugin support Jan 04 17:42:23 nbd r35014 trunk/package/mac80211/Makefile * mac80211: add dependency of PACKAGE_MAC80211_DEBUGFS Jan 04 17:53:23 IPT_BUILTIN += $(EBTALTES_IP6-y) Jan 04 17:53:28 EBTALTES? nice one >_> Jan 04 17:53:47 (include/netfilter.mk) Jan 04 17:56:42 build #113 of ar71xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ar71xx/builds/113 Jan 04 18:12:03 build #117 of ixp4xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/117 Jan 04 19:06:25 * strasidlo is finishing asterisk11 package for OpenWrt Jan 04 20:20:12 build #113 of au1000 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/113 Jan 04 21:56:07 build #105 of adm5120 is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/105 Jan 04 23:30:27 kaloz: ping Jan 05 00:03:18 build #141 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/141 Jan 05 00:21:05 blogic r35015 trunk/ target/linux/generic/patches-3.3/270-regmap_dynamic.patch target/linux/generic/patches-3.7/259-regmap_dynamic.patch package/kernel/modules/other.mk * [kernel] fix regmap debloat patch Jan 05 01:12:29 build #101 of mpc52xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/101 Jan 05 02:21:51 build #141 of brcm47xx is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/141 Jan 05 02:28:40 build #130 of rb532 is complete: Failure [failed shell_11] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/130 Jan 05 02:32:20 build #130 of ppc44x is complete: Failure [failed shell_11] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/130 Jan 05 02:50:34 build #134 of lantiq is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/lantiq/builds/134 **** ENDING LOGGING AT Sat Jan 05 02:59:59 2013