**** BEGIN LOGGING AT Tue Mar 13 02:59:58 2012
Mar 13 04:46:43 <builder_> build #103 of mpc52xx is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/103
Mar 13 08:26:26 <CIA-54> nbd * r30915 /trunk/package/mac80211/patches/ (3 files): ath9k: remove stuck beacon detection optimization, it may be unreliable on some hardware
**** BEGIN LOGGING AT Tue Mar 13 09:48:19 2012
Mar 13 09:48:55 <CIA-54> nbd * r30916 /trunk/package/base-files/files/etc/init.d/boot: base-files: remove old coldplugging hacks that are no longer necessary since the new main hotplug2 instance calls udevtrigger as well
Mar 13 10:04:42 <jow_laptop> philipp64|laptop: there should be nothing wrong with using uclibc++ under eglibc
Mar 13 10:04:50 <jow_laptop> thee isn't an eglibc++ after all
Mar 13 10:34:13 <builder_> build #121 of xburst is complete: Failure [failed compile_10]  Build details are at http://buildbot.openwrt.org:8010/builders/xburst/builds/121
Mar 13 10:37:10 <CIA-54> nbd * r30917 /trunk/target/linux/generic/ (2 files in 2 dirs): kernel: bypass all netfilter hooks if the sysctls for that functionality have been disabled - eliminates the overhead of enabling CONFIG_BRIDGE_NETFILTER in the kernel config
Mar 13 11:21:45 <dape> !seen jow_laptop
Mar 13 11:21:56 <dape> meh, worth to try
Mar 13 11:49:44 <CIA-54> nbd * r30918 /trunk/package/mac80211/patches/565-ath9k_optimize_queue_wake.patch: ath9k: optimize waking tx queues, slightly improves performance under load
Mar 13 12:48:27 <builder_> build #136 of ps3 is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ps3/builds/136
Mar 13 12:54:03 <stintel> heh. openwrt still builds for ps3?
Mar 13 12:54:08 <stintel> otheros is long gone, no ?
Mar 13 12:55:55 <CIA-54> jow * r30919 /trunk/package/base-files-network/ (Makefile files/lib/network/config.sh): [package] base-files-network: configure vlan interface name type immediately before the vconfig add calls, fixes race condition (#11127)
Mar 13 12:58:28 <dape> jow_laptop ping
Mar 13 12:58:38 <jow_laptop> .-
Mar 13 12:59:17 <dape> jow_laptop i "cracked" the dir615 h1 last night :) they use a /sbin/config_term kind of bin login, found the password sn2450 with strings in the gpl source binary
Mar 13 13:00:00 <dape> had wide dhcpv6 misconfigured, pppd with ipv6 compiled but not enabled, old radvd and all kind of weird components for web management
Mar 13 13:00:42 <dape> i know it sounds strange "gpl source binary" but in the tarball from dlink's site wasnt any source for /sbin/config_term , just the binary...
Mar 13 14:37:10 <CIA-54> florian * r30920 /trunk/target/linux/brcm63xx/ (89 files in 3 dirs): [brcm63xx] remove udc patch since it is unused
Mar 13 16:22:40 <philipp64|laptop> jow_laptop: well, it's failing to compile whereas it wasn't before.
Mar 13 16:23:45 <philipp64|laptop> see http://fpaste.org/qqUO/
Mar 13 16:27:07 <[florian]> it always failed building that way for me
Mar 13 16:40:19 <philipp64|laptop> ok, so how do I figure out what requires it? because it's not explicitly selected.
Mar 13 16:56:22 <[florian]> reading tmp/.package_deps is probably the easiest way
Mar 13 17:30:19 <tripolar> ping jow_laptop
Mar 13 17:30:21 <tripolar> hi
Mar 13 17:31:00 <tripolar> jow the patches r30916 and r30919 seem to introduce an error - my vlan seems to be created in a wrong way and therefor my router isnt reachable
Mar 13 17:31:21 <tripolar> reverting both commits fixes the problem and everything works as expected
Mar 13 17:31:49 <tripolar> ping nbd
Mar 13 17:33:21 <tripolar> jow_laptop: the patches r30916 and r30919 seem to introduce an error - my vlan seems to be created in a wrong way and therefor my router isnt reachable
Mar 13 17:39:37 <jow_laptop> details?
Mar 13 17:41:36 <tripolar> i don't have access to the device but it seems the is a vlan config error - im unsing brcm47xx and a wl500gp
Mar 13 17:41:51 <tripolar> i mean i dont have a serial port connected to the device
Mar 13 17:42:19 <tripolar> if i compile a image with both commits applied i can't reach the device
Mar 13 17:42:23 <tripolar> its not pingable
Mar 13 17:42:42 <tripolar> when i revert both it works as it should - the device is reachable and pingable
Mar 13 17:44:04 <tripolar> i could replace both reverted files with a original ones and log to a file
Mar 13 17:44:10 <tripolar> but what should i log?
Mar 13 17:44:39 <jow_laptop> try to only revert the first part of r30916
Mar 13 17:44:45 <jow_laptop> the rest is unrelated to your problem
Mar 13 17:46:05 <tripolar> jow_laptop: okay
Mar 13 17:53:45 <tripolar> jow_laptop: you where right - reverting only first part of r30916 fixed it
Mar 13 17:53:50 <tripolar> jow_laptop: you were right - reverting only first part of r30916 fixed it
Mar 13 17:59:48 <philipp64|laptop> jow_laptop: 30919 this looks familiar.
Mar 13 17:59:58 <CIA-54> juhosg * r30921 /trunk/target/linux/ar71xx/ (2 files in 2 dirs): ar71xx: fix a typo in ar71xx_regs.h
Mar 13 18:00:00 <CIA-54> juhosg * r30922 /trunk/target/linux/ar71xx/files/arch/mips/ath79/dev-eth.c: ar71xx: reset the switch on AR934x before ethernet device registration
Mar 13 18:00:01 <CIA-54> juhosg * r30924 /trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_ar7240.c: ar71xx: ag71xx: poll PHY status of all available switch ports
Mar 13 18:00:03 <CIA-54> juhosg * r30923 /trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_ar7240.c: ar71xx: ag71xx: allow to use port 5 of the AR934x built-in switch
Mar 13 18:00:04 <CIA-54> juhosg * r30925 /trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_ar7240.c: ar71xx: ag71xx: start aneg on switch PHYs after reset
Mar 13 18:00:05 <CIA-54> juhosg * r30926 /trunk/target/linux/ar71xx/files/drivers/net/ethernet/atheros/ag71xx/ag71xx_ar7240.c: ar71xx: ag71xx: fix ARP frame reception on AR934x switch ports
Mar 13 18:00:14 <CIA-54> juhosg * r30927 /trunk/target/Makefile: imagebuilder: fix parallel building (closes #11117)
Mar 13 18:00:30 <philipp64|laptop> looking at the kirkwood code by the way, and wondering why the "root" => "rootfs" patch is required.
Mar 13 18:01:37 <jow_laptop> because its easier to change root->rootfs in one place than to patch scripts ina dozen places
Mar 13 18:03:37 <CIA-54> juhosg * r30928 /trunk/target/linux/generic/files/drivers/net/phy/ar8216.c: generic: ar8216: merge multiple if statements in ar8216_hw_apply
Mar 13 18:04:43 <jow_laptop> tripolar: whats the result of "find /sys/ -name uevent | grep eth0" on your model?
Mar 13 18:05:14 <tripolar>  /sys/devices/virtual/net/eth0.0/uevent
Mar 13 18:05:23 <jow_laptop> hm as expected
Mar 13 18:05:24 <tripolar>  /sys/devices/virtual/net/eth0.1/uevent
Mar 13 18:05:30 <tripolar>  /sys/devices/ssb0:0/net/eth0/uevent
Mar 13 18:05:32 <tripolar> this three
Mar 13 18:06:28 <philipp64|laptop> ok, but I was talking to the archlinux folks last night. they sent me a patch that adds 4 new platforms and builds under 3.1.10.
Mar 13 18:06:49 <philipp64|laptop> maybe configurable scripts wouldn't be so bad. oh, well.
Mar 13 18:07:12 <philipp64|laptop> they also pointed out that a lot of kirkwood support got upstreamed.
Mar 13 18:07:28 <jow_laptop> tripolar: can you show me the output of "udevtrigger -v -n" ?
Mar 13 18:08:04 <tripolar> http://nopaste.info/23ef4f4fd0.html
Mar 13 18:08:12 <tripolar> @jow_laptop
Mar 13 18:08:46 <jow_laptop> philipp64|laptop: variable mtd names also caused quite some mess on atheros where kernel can be named "kernel", "vmlinuz", "vmlinux.bin.l7", "linux", "image" or similar
Mar 13 18:09:04 <jow_laptop> I prefer a one line patch any time over that
Mar 13 18:10:08 <jow_laptop> tripolar: it would be interesting to write this log at the time the now restored /proc/net/dev coldplugging happens
Mar 13 18:10:18 <jow_laptop> tripolar: preferably before
Mar 13 18:10:36 <jow_laptop> tripolar: I want to see whether udevtrigger would trigger eth0
Mar 13 18:11:06 <tripolar> so where should i log?
Mar 13 18:11:21 <tripolar> i can log to a hd and read the file even when i dont have access to the device
Mar 13 18:11:54 <jow_laptop> failsafe should still work, despite the broken hotplug
Mar 13 18:12:36 <jow_laptop> but you could also try to put "udevtrigger -v n 2>&1 >/tmp/trigger.log" before "for iface in $(awk -F: '/:/ {print $1}' /proc/net/dev); do" for now
Mar 13 18:21:34 <tripolar> jow_laptop: http://nopaste.info/24e4e00ce6.html
Mar 13 18:21:53 <jow_laptop> hm
Mar 13 18:22:25 <jow_laptop> I need to debug this myself
Mar 13 18:23:07 <jow_laptop> according to your log it should trigger vlans just fine
Mar 13 18:23:46 <jow_laptop> hope I can reproduce it on a wrt54gs
Mar 13 18:23:52 <tripolar> jow_laptop: this two lines are missing /devices/virtual/net/eth0.0
Mar 13 18:24:01 <tripolar>  /devices/virtual/net/eth0.1
Mar 13 18:34:34 <philipp64|laptop> [florian]: drawing a blank about the uclibc++ thing. sigh.
Mar 13 19:26:33 <tripolar> jow_laptop: could you reproduce it?
Mar 13 19:34:31 <btsimonh> blogic: how are you doing?
Mar 13 19:36:20 <blogic> fine
Mar 13 19:36:28 <blogic> i have added bthh to my local tree
Mar 13 19:36:29 <blogic> ;)
Mar 13 19:36:33 <blogic> will test it tomorrow
Mar 13 19:36:38 <blogic> but it builds
Mar 13 19:36:46 <blogic> needs to go over the wifi stuff
Mar 13 19:36:51 <blogic> and check the nand stuff
Mar 13 19:47:18 <philipp64|laptop> [florian]: what was your workaround to the uclibc++ build failure?
Mar 13 19:51:24 <jow_laptop> tripolar: not yet
Mar 13 19:52:19 <jow_laptop> tripolar: vlan interfaces are created so the problem is not the udev trigger
Mar 13 19:52:30 <jow_laptop> tripolar: I suppose the switch setup is broken
Mar 13 19:53:09 <tripolar> but why does it work with the  first part of r30916 reverted?
Mar 13 19:53:36 <jow_laptop> probebly a coincidence hinting at a larger problem
Mar 13 20:00:31 <tripolar> okay
Mar 13 20:11:45 <CIA-54> juhosg * r30929 /packages/devel/4th/ (. Makefile):
Mar 13 20:11:45 <CIA-54> packages/4th: forth compiler
Mar 13 20:11:45 <CIA-54> [juhosg: remove non-english comments]
Mar 13 20:11:45 <CIA-54> Signed-off-by: Xiangfu <xiangfu.z@gmail.com>
Mar 13 20:11:46 <CIA-54> juhosg * r30930 /packages/utils/byobu/ (. Makefile):
Mar 13 20:11:47 <CIA-54> packages/byobu: decorative, multi-panel screens terminal
Mar 13 20:11:47 <CIA-54> [juhosg: reformat package description]
Mar 13 20:11:47 <CIA-54> Signed-off-by: Xiangfu <xiangfu.z@gmail.com>
Mar 13 20:11:48 <CIA-54> juhosg * r30931 /packages/utils/calcurse/ (. Makefile):
Mar 13 20:11:48 <CIA-54> packages/calcurse: fullscreen curses based calendar
Mar 13 20:11:49 <CIA-54> Signed-off-by: Xiangfu <xiangfu.z@gmail.com>
Mar 13 20:11:49 <CIA-54> juhosg * r30932 /packages/utils/taskwarrior/ (. Makefile):
Mar 13 20:11:50 <CIA-54> packages/taskwarrior: command line task manager
Mar 13 20:11:50 <CIA-54> Signed-off-by: Xiangfu <xiangfu.z@gmail.com>
Mar 13 20:21:17 <jow_laptop> tripolar: yes indeed, the interfaces are all set up fine but the vlans in the switch are not
Mar 13 20:21:44 <jow_laptop> in my case it only activates port 0
Mar 13 20:39:09 <jow_laptop> yes, looks like some simple race condition
Mar 13 20:39:27 <jow_laptop> I did put the switch setup into a subshell for debug purposes and it works fine now
Mar 13 20:48:11 <tripolar> okay - weired
Mar 13 21:04:25 <jow_laptop> a sleep 5 works too
Mar 13 21:04:27 <jow_laptop> hmm
Mar 13 21:11:57 <CIA-54> jow * r30936 /trunk/package/uhttpd/src/Makefile: [package] uhttpd: reorder compiler flags to fix native build on Ubuntu 11.x
Mar 13 21:34:05 <jow_laptop> tripolar: still there?
Mar 13 21:34:12 <tripolar> jow_laptop: yes
Mar 13 21:34:43 <tripolar> do you have a patch i could try?
Mar 13 21:34:52 <jow_laptop> yes
Mar 13 21:35:34 <jow_laptop> http://pastebin.com/tN11ZUBM
Mar 13 21:35:49 <jow_laptop> you can do it manually on your device in /lib/network/switch.sh
Mar 13 21:36:54 <tripolar> and remove the other part in init.d/boot?
Mar 13 21:37:01 <jow_laptop> yes
Mar 13 21:38:42 <tripolar> okay done i reboot now
Mar 13 21:44:01 <tripolar> jow_laptop: it works
Mar 13 21:46:07 <CIA-54> jow * r30937 /trunk/package/switch/ (Makefile files/switch.sh):
Mar 13 21:46:07 <CIA-54> [package] switch: the hardware must be up while configuring
Mar 13 21:46:07 <CIA-54> Since the removal of the interface coldplug hack the timing changed in such a way that eth0 is not up during switch setup,
Mar 13 21:46:07 <CIA-54> leading to an inaccesible device. Bring up the switch interface before configuring it to mitigate the problem.
Mar 13 21:46:12 <jow_laptop> great
Mar 13 21:49:23 <tripolar> thanks for the fix
Mar 13 21:50:17 <jow_laptop> yw, thanks for testing
Mar 13 21:51:01 <tripolar> yw :)
Mar 13 21:52:38 <CIA-54> jow * r30938 /trunk/package/firewall/ (Makefile files/lib/core_interface.sh): [package] firewall: fix fw__uci_state_del() procedure (#11132)
**** BEGIN LOGGING AT Tue Mar 13 22:04:38 2012
Mar 13 22:37:35 <tripolar> jow_laptop: i just sent to patch to the ml - that updates openssl to 1.0.0h. could you please apply it
Mar 13 22:39:04 <jow_laptop> already on it
Mar 13 22:39:17 <tripolar> :)
Mar 13 22:39:19 <tripolar> thanks
Mar 13 22:40:15 <CIA-54> jow * r30939 /trunk/package/openssl/ (4 files in 2 dirs):
Mar 13 22:40:15 <CIA-54> openssl 1.0.0h update patch
Mar 13 22:40:15 <CIA-54> Hi
Mar 13 22:40:15 <CIA-54> this patch updates openssl to 1.0.0h and refreshes all patches.
Mar 13 22:40:15 <CIA-54> This fixes CVE-2012-0884.
Mar 13 22:40:16 <CIA-54> Signed-off-by: Peter Wagner <tripolar@gmx.at>
Mar 14 00:03:51 * swalker sighs at xiangfu's packages
Mar 14 00:05:24 <Chocks> yes
Mar 14 00:05:27 <Chocks> crazy stuff
Mar 14 00:06:27 <philipp64|laptop> anyone using xt_geoip or understand how it works?
**** BEGIN LOGGING AT Wed Mar 14 00:54:05 2012
Mar 14 01:00:23 <philipp64|laptop> jow_laptop: ping
Mar 14 02:46:48 <builder_> build #119 of kirkwood is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/kirkwood/builds/119
**** ENDING LOGGING AT Wed Mar 14 02:59:58 2012