**** BEGIN LOGGING AT Wed Mar 06 02:59:58 2013 Mar 06 04:08:26 build #206 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/206 Mar 06 06:35:22 build #199 of ppc40x is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/199 Mar 06 06:40:31 build #186 of rb532 is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/rb532/builds/186 Mar 06 06:44:54 build #186 of ppc44x is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/186 Mar 06 07:15:15 * russell-- has iproute2-v3.8.0 built against HEAD Mar 06 07:21:34 build #191 of uml is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/191 Mar 06 08:00:03 build #188 of x86 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/x86/builds/188 Mar 06 08:31:16 build #176 of sibyte is complete: Failure [failed shell_12] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/176 Mar 06 10:59:03 ooh, interesting, just flashed a test image onto a wzr600dhp, iptables is segfaulting on me Mar 06 11:00:32 r35893 Mar 06 11:07:58 cyrusff: is r35892 yours? Mar 06 11:08:13 russell--: yes i'm currently investigating what is wrong Mar 06 11:08:20 cool Mar 06 11:09:02 seems there is a stupid heisenbug, everything works fine when i disable stripping for iptables, but when i enable stripping (default) it segfaults Mar 06 11:09:13 at least on ar71xx Mar 06 11:09:16 awesome! Mar 06 11:10:47 Hello! russell--, was it you whom I told about WNDR3800 crashing on IPv6 ping? Mar 06 11:12:01 cyrusff: http://sigquit.wordpress.com/2008/09/15/debugging-programs-with-a-external-symbol-file/ Mar 06 11:12:06 BasicXP: yes Mar 06 11:12:28 The threshold is somewhere between 1200 and 1250 bytes Mar 06 11:12:58 The thing is it used to crash at 1200, now it doesn't. The only thing I changed in the configuration is removed the "wan6" interface that is being created by default. Mar 06 11:13:05 * russell-- hasn't cleared a spot on his workbench yet Mar 06 11:13:25 * BasicXP nods Mar 06 11:13:55 i did ping6 at other similar ar71xx's and didn't replicate the behavior Mar 06 11:14:06 And pinging the router itself doesn't crash it at highest values. Mar 06 11:14:51 have you tried updating source and replicating? Mar 06 11:15:22 Right now I have revision 35863 installed. Mar 06 11:15:52 or starting from a vanilla configuration? Mar 06 11:16:29 that is, rule out your configuration as the cause Mar 06 11:17:04 The build configuration, you mean? Mar 06 11:17:11 no, on the device Mar 06 11:17:55 reconfigure it from scratch? I mean, I won't be able to access IPv6 without some configurations Mar 06 11:18:21 right, but make them as minimal as possible Mar 06 11:19:18 Okay, will try that. Mar 06 11:19:24 That brings another problem, by the way. Mar 06 11:19:24 it's panicking? Mar 06 11:19:39 How do I determine? It just reboots. No crashlog. Mar 06 11:19:48 (/me tries to reload context ;-) Mar 06 11:20:00 BasicXP: what crashes? an application or the whole kernel when you receive the ping? Mar 06 11:20:31 when he ping6's a large packet to an remote host Mar 06 11:20:46 The whole router reboots Mar 06 11:20:46 the wndr3800 reboots Mar 06 11:20:49 wow Mar 06 11:22:59 * russell-- is about to turn in here on us west cost, hoping cyrusff figures out iptables problem by my waking time ;-) Mar 06 11:23:36 I guess I'll try with a fresh config Mar 06 11:23:49 Though I have to say I have an issue with PPPoE hotplug script Mar 06 11:23:55 hehe if i cannot figure it out right now, i will revert and do it later Mar 06 11:24:19 If I manually do /etc/init.d/network restart, the script doesn't seem to kill the previous pppd instance, yet it should Mar 06 11:24:27 So more and more ppp-interfaces are created Mar 06 11:25:16 kind of a big jump from 1.4.10 to 1.4.18, not surprising there is some carnage Mar 06 11:25:28 yeah but the problem is toolchain-related Mar 06 11:25:34 ah Mar 06 11:25:40 or something like that Mar 06 11:25:55 if i disable stripping the binaries it works Mar 06 11:26:08 which should not happen Mar 06 11:26:17 right Mar 06 11:27:30 BasicXP: if you build a newer version, be sure to rebuild your kernel. "make ... target/linux/clean world" Mar 06 11:28:09 does "make clean" not do that by default? Mar 06 11:28:18 * russell-- has stepped on that land mine a few times Mar 06 11:28:22 oh, maybe. Mar 06 11:28:37 I always run make clean before updating the source tree Mar 06 11:28:54 * russell-- doesn't use make clean, or at least not often Mar 06 11:30:21 And regarding PPPoE: http://pastebin.com/Q0t3HvL8 Mar 06 11:30:41 Wonder why on manual network restart the ppp interface is not deleted Mar 06 11:32:57 should that be [ "$INTERFACE" = wan2 ] ? Mar 06 11:33:08 * russell-- doesn't use pppoe Mar 06 11:34:44 * russell-- would try replicating the ipv6 bug without the multi-wan configuration Mar 06 11:34:45 No, it should be "wan", since that's the actual WAN interface that is set up to use DHCP Mar 06 11:34:53 We have dual access Mar 06 11:35:02 I will try it with plain PPPoE Mar 06 11:37:44 I see you need to go, russell--, so have a good night :) Mar 06 11:39:51 and cyrusff, you think it's better to disable stripping for now since iptables segfaults on ar71xx? Mar 06 11:41:00 no, i will try debugging the issue (for now it seems that some .so-constructors run before some other init functions that are required to initialize some structs) Mar 06 11:41:10 and if not revert to 1.4.10 until the issue can be resolved Mar 06 12:56:23 cyrus r35894 trunk/package/network/ (11 files in 2 dirs) Mar 06 12:56:23 Revert "iptables: update to 1.4.18" due to toolchain-issue: Mar 06 12:56:23 binaries cause segfaults when stripped on ar71xx Mar 06 12:59:30 russell--: ^ i hope i can find some time later this week to debug this mess Mar 06 13:29:52 mirko r35895 trunk/README * Update README file Mar 06 13:37:43 mirko: ping Mar 06 13:57:55 wbx: pong Mar 06 15:17:41 mirko: just wanted to note that the README might need some more changes. pkg-config is not needed, it is in tools/ Mar 06 15:18:03 mirko: but I prepare some more changes and sent it with the bsd patches... Mar 06 15:31:40 wbx: i know - was just a hot-fix because some user struggled over this once again Mar 06 15:32:47 mirko: okay. Mar 06 16:29:13 sigh Mar 06 16:29:42 going to have to redo image making code a bit i guess Mar 06 16:30:42 squashfs sig invalidates the CRC Mar 06 16:30:55 erh, squashfs split sig + jffs2 Mar 06 17:05:58 cyrus r35896 trunk/package/network/ (12 files in 2 dirs) * iptables: redo update to 1.4.18 with old linking-behaviour Mar 06 18:25:43 cyrusff: does that fix it? Mar 06 18:29:07 russell--: the segfaults? yes. Mar 06 18:29:49 cools Mar 06 18:30:24 * russell-- building here Mar 06 18:32:03 i had it running on a tl-wdr3600 after a make clean; make Mar 06 18:34:44 * russell-- doing a make package/iptables/clean world Mar 06 18:54:56 I need a jtag cable because I bricked on of my devices, should this work with a WRT610N (Broadcom BCM4705): http://www.ebay.de/itm/SainSmart-USB-Blaster-Programmer-Cable-For-FPGA-CPLD-JTAG-Development-Board-/320987021787?pt=DE_Technik_Computerzubeh%C3%B6r_Kabel_Adapter&hash=item4abc5141db Mar 06 20:34:07 Hauke: you should figure out which software you are planning to use, then make sure the software supports the jtag interface device Mar 06 20:35:04 * russell-- has never actually done any mips jtag at all, just a little arm jtag (with openocd) Mar 06 20:39:15 russell--: thanks for the info Mar 06 23:26:58 luka r35897 packages/net/strongswan/Makefile * [packages] strongswan: update to version 5.0.2 **** ENDING LOGGING AT Thu Mar 07 02:59:58 2013