**** BEGIN LOGGING AT Thu Jul 11 02:59:58 2013
Jul 11 05:27:10 <cyrusff> raorn: did you use ip6assign instead of ip6addr?
Jul 11 05:29:28 <cyrusff> as stated in the wiki-page?
Jul 11 05:40:22 <builder> build #305 of ppc40x is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/305
Jul 11 06:17:15 <The_Lizard> blogic: I don't know if you managed to try out the EASY80920, my computer decided to reboot so I don't know if you posted anything about it...
Jul 11 06:54:09 <builder> build #300 of uml is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/300
Jul 11 06:55:34 <__trine> has something changed with mac80211 I notice there is no makefile or any of the usual patches in my build environment
Jul 11 06:58:00 <builder> build #244 of iop32x is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/244
Jul 11 07:11:29 <blogic> __trine: it got moved
Jul 11 07:11:38 <blogic> package/kernel/mac80211
Jul 11 07:11:56 <__trine> blogic, thanks
Jul 11 07:38:21 <raorn> cyrusff: my provider gives me /64 prefix
Jul 11 07:38:33 <cyrusff> so?
Jul 11 07:39:04 <cyrusff> is that problematic?
Jul 11 07:39:53 <raorn> i don't know. i'll try it then
Jul 11 07:40:20 <cyrusff> just follow the instructions in the wiki
Jul 11 07:40:28 <cyrusff> doesn't matter if its /64
Jul 11 07:49:29 <raorn> hm....
Jul 11 07:50:00 <raorn> cyrusff: what's the difference? i got the same $prefix::1/64 on br-lan
Jul 11 07:50:29 <cyrusff> the difference is that you don't get an icmp error 5 if the clients use the address ;)
Jul 11 07:51:14 <raorn> i'm an engineer, i want to know why ;-)
Jul 11 07:51:37 <cyrusff> ok, the long story then... ;)
Jul 11 07:52:24 <cyrusff> we have source-adress specific routing in-place as there are several rfcs that ipv6-routers shoudl follow (6204 being the most important one)
Jul 11 07:52:56 <cyrusff> and one of the requirements is that if clients use source addresses from disallowed / unknwon prefixes they should be denied with icmpv6 code 5
Jul 11 07:54:03 <cyrusff> so by default we block everything with code 5 and only route traffic which comes from a prefix assigned through the ip6assign/ip6hint/ip6class logic
Jul 11 07:54:39 <cyrusff> this is mainly that prefix-changes from the ISP or multiple IPv6 uplinks are handled correctly
Jul 11 07:54:53 <raorn> so, ip6addr should only be used on static wan interface
Jul 11 07:55:10 <cyrusff> yes you should only use it for uplink-interfaces
Jul 11 07:59:01 <raorn> well, i love new openwrt
Jul 11 07:59:42 <raorn> btw, i've found another nat64 implementation - https://github.com/NICMx/NAT64
Jul 11 07:59:57 <raorn> this time it's kernel-space
Jul 11 08:01:03 <cyrusff> i see
Jul 11 08:01:15 <cyrusff> however i think we still need a suitable dns64
Jul 11 08:07:37 <raorn> bind works fine
Jul 11 08:09:06 <cyrusff> yeah except that its huge
Jul 11 08:12:27 <stintel> :D
Jul 11 08:20:17 <builder> build #288 of ar7 is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/288
Jul 11 08:27:00 <raorn> cyrusff: how exactly you "block everything"? definitely not with ip6tables
Jul 11 08:27:19 <cyrusff> ip policy rules
Jul 11 08:27:26 <cyrusff> see ip -6 rule
Jul 11 08:27:33 <cyrusff> if you have ip installed
Jul 11 08:30:53 <raorn> i see...
Jul 11 08:49:14 <stintel> russell--: I configured my rsyslog to also listen on TCP/514 for now
Jul 11 08:50:57 <stintel> russell--: did you mention something about dropbear refusing connections ? looks like I am seeing it as well
Jul 11 08:51:24 <stintel> Thu Jul 11 10:43:33 2013 security.info dropbear[1739]: Premature exit: Terminated by signal
Jul 11 08:56:09 <raorn> what's the point of having fdXX:: address on br-lan?
Jul 11 08:56:59 <cyrusff> local ipv6 addresses in case there is no wan-connection
Jul 11 08:57:10 <cyrusff> but you can disable them if you like
Jul 11 08:58:03 <raorn> how? clear out the IPv6 ULA-Prefix?
Jul 11 08:58:12 <cyrusff> yes for example
Jul 11 08:58:33 <cyrusff> or using ip6class (see wiki)
Jul 11 09:01:26 <KanjiMonster> raorn: think about that case: assume your pc/device is ipv6 only, how do you connect to the router/configure it before it is configured and got a prefix from upstream (or if your internet connection is down)
Jul 11 09:06:03 <raorn> KanjiMonster: before 6relayd i had radvd with pre-defined configuration
Jul 11 09:07:42 <KanjiMonster> raorn: but that means that you already connected to and configured the router
Jul 11 09:08:51 <KanjiMonster> maybe you don't need it, but it doesn't mean that there is no reason for others to have it (also iirc rfc6204 defines that it should be present by default)
Jul 11 09:28:07 <raorn> i'll keep it. sounds reasonable
Jul 11 09:33:06 <cyrusff> yeah one of the main purposes of the new ipv6-stack is that you don't have enter any prefixes manually
Jul 11 09:33:17 <cyrusff> which would be bad with ipv6 pd
Jul 11 09:36:04 <stintel> hmmm. logread also doesn't seem to send the hostname when doing remote syslog
Jul 11 09:36:10 <stintel> so my logfile ends up "Thu.log"
Jul 11 09:36:10 <stintel> meh
Jul 11 09:36:37 <jow_laptop> I believe the not-send-hostname thing was a limitation in busybox too
Jul 11 09:37:41 <cyrusff> shouldn't be that problematic to write a patch i guess
Jul 11 09:38:22 <stintel> jow_laptop: but it did send IP address back then iirc
Jul 11 09:38:42 <jow_laptop> stintel: ah ok, I see
Jul 11 09:38:46 <stintel> Jul  2 14:07:25 192.168.50.254 ipsec_starter[2719]: ipsec starter stopped
Jul 11 09:38:46 <stintel> Jul 11 11:37:49 Thu Jul 11 11:37:49 2013 daemon.alert logread[2324]: Logread connected to 192.168.50.16:514
Jul 11 09:38:52 <jow_laptop> well my point was that we now have the chance to actually implement the hostname
Jul 11 09:39:02 <stintel> that was old vs new behaviour
Jul 11 09:41:54 <stintel> seems I have a dupe timestamp now :P
Jul 11 09:42:22 <stintel> probably 1 local and 1 remote
Jul 11 09:43:52 <raorn> does miniupnpd support ipv6? of course there is no nat in ipv6 (yet?), but it could just allow incoming connections...
Jul 11 09:44:12 <jow_laptop> it supports ipv6
Jul 11 09:44:20 <jow_laptop> at least listening for ipv6 multicast
Jul 11 09:44:29 <jow_laptop> no iptables support yet I believe
Jul 11 09:45:01 <stintel> ah sweet, my ISP enabled native v6 on my new connection (I moved)
Jul 11 09:45:19 <stintel> it's in internal trial for now, lucky I am currently working there :)
Jul 11 09:45:20 <jow_laptop> my isp still wonders what ipv6 is
Jul 11 09:53:59 <russell--> stintel: yes, but no idea why
Jul 11 09:54:05 <russell--> re dropbear
Jul 11 10:06:58 * russell-- is seeing logread start (there is a /var/run/logread.1.pid but remote syslog'ing is broken after boot, works after /etc/init.d/log restart
Jul 11 10:07:32 <russell--> oh, wait, confirming that ...
Jul 11 10:10:41 <russell--> yes, that was the case this time anyway
Jul 11 10:11:02 <russell--> no output of any boot messages
Jul 11 10:11:23 <russell--> and remote syslogs receive no "logger" messages until i restart
Jul 11 10:12:07 * russell-- says "logger hi!", show in local syslog (e.g. output of local logread) but not in remote syslogs until i restart
Jul 11 10:46:51 <builder> build #281 of avr32 is complete: Failure [failed shell]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/281
Jul 11 11:03:10 <tripolar> blogic: i fixed the root delay. i sent a patch to the ml. could you apply my other patch to a sent a few days ago?
Jul 11 11:03:52 <tripolar> to a/i
Jul 11 13:41:53 <OpenWrt-Commits> luka r37238 packages/net/snort/Makefile * [packages] snort: add libuuid as dependency
Jul 11 13:48:47 <builder> build #272 of ixp4xx is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/272
Jul 11 14:09:18 <tripolar> how do i add a link to the a rev [x] when i close a bug and say fixed in [x]?
Jul 11 14:09:26 <tripolar> just []?
Jul 11 14:09:51 <jow_laptop> Just write "r12345"
Jul 11 14:09:59 <tripolar> jow_laptop: thanks
Jul 11 14:10:06 <jow_laptop> trac replaces any /r\d+/ with a link to said rev
Jul 11 14:10:42 <tripolar> jow_laptop: with ""?
Jul 11 14:12:27 <luka> tripolar: no, without ""
Jul 11 14:12:27 * russell-- has made a thorough study and determined unequivocally that 3 hours is not enough sleep
Jul 11 14:12:57 <tripolar> russell--: you will get ill with less then 7 hours for a longer period of time
Jul 11 14:13:04 <tripolar> then=than
Jul 11 14:13:14 * russell-- feels ill already ;-)
Jul 11 14:13:32 <russell--> nap scheduled
Jul 11 14:15:37 <tripolar> jow_laptop and how do i point in a commit to a bug?
Jul 11 14:15:43 <tripolar> also r?
Jul 11 14:16:50 <stintel> I think just #1234
Jul 11 14:16:53 * russell-- vaguely remembers #n ... but don't rely on that
Jul 11 14:16:57 <stintel> isn't there a sandbox ? :)
Jul 11 14:17:33 <russell--> there is also lmgtfy ;-)
Jul 11 14:18:53 <OpenWrt-Commits> tripolar r37239 packages/net/ git/Makefile git/patches/200-disable_fasthash.patch
Jul 11 14:18:53 <OpenWrt-Commits> [packages] git: get rid of XDL_FAST_HASH, it does NOT work on big-endian systems,
Jul 11 14:18:53 <OpenWrt-Commits> and uname is no reliable way to determine which arch will run git
Jul 11 14:18:53 <OpenWrt-Commits> closes #13611
Jul 11 14:19:06 <russell--> tripolar: http://trac.edgewall.org/wiki/TracCrossReferences
Jul 11 14:20:20 <tripolar> russell--: i guessed right ;)
Jul 11 14:20:50 <tripolar> unbelieveable git uses uname to determine the arch on compile time, thought linux people would think about cross compiling
Jul 11 14:21:03 <tripolar> at least linus ;)
Jul 11 14:21:31 <groz1> remember the history, git came about essentially 'overnight' because of problems in the world
Jul 11 14:21:50 <groz1> so it was rushed initially, and, 'quick and dirty' decisions to rush it, probably still live in there
Jul 11 14:21:52 <groz1> years later
Jul 11 14:22:33 <tripolar> at they mention the big-endian problem
Jul 11 14:22:39 <tripolar> at least they mention the big-endian problem
Jul 11 14:46:35 <builder> build #247 of ep93xx is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/247
Jul 11 14:50:54 <builder> build #216 of octeon is complete: Failure [failed shell_12]  Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/216
Jul 11 16:57:59 <builder> build #245 of mcs814x is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/245
Jul 11 17:02:00 <OpenWrt-Commits> blogic r37240 trunk/package/base-files/files/etc/init.d/usb * base-file: usbfs is no longer mounted under /proc/bus/usb
Jul 11 17:02:04 <OpenWrt-Commits> blogic r37241 trunk/package/base-files/files/etc/rc.common * base-files: run readlink on initscript name before opening services
Jul 11 17:02:09 <OpenWrt-Commits> blogic r37242 trunk/package/utils/busybox/files/telnet * busybox: convert telnet init script to procd
Jul 11 17:02:13 <OpenWrt-Commits> blogic r37243 trunk/package/utils/busybox/files/cron * busybox: convert crond init script to procd
Jul 11 17:02:18 <OpenWrt-Commits> blogic r37244 trunk/package/base-files/files/etc/init.d/sysntpd * base-files: convert sysntpd init script to procd
Jul 11 17:02:23 <OpenWrt-Commits> blogic r37245 trunk/package/network/services/dropbear/files/dropbear.init * dropbear: register a config.change trigger
Jul 11 17:02:27 <OpenWrt-Commits> blogic r37246 trunk/package/system/ubox/Makefile * ubox: upgrade to latest version
Jul 11 17:02:32 <OpenWrt-Commits> blogic r37247 trunk/package/system/procd/files/procd.sh * procd: the delete ubus call was passed the wrong field name for services
Jul 11 17:02:37 <OpenWrt-Commits> blogic r37248 trunk/package/ (5 files in 4 dirs) * procd: add a small script that handles config reloads until configd is ready
Jul 11 17:02:42 <OpenWrt-Commits> blogic r37249 trunk/package/system/procd/files/log.init * procd: add proto and trigger support to the /etc/init.d/log
Jul 11 17:02:46 <OpenWrt-Commits> blogic r37250 trunk/package/system/procd/Makefile * procd: update to latest git version
Jul 11 17:02:51 <OpenWrt-Commits> blogic r37251 trunk/target/linux/ar71xx/modules.mk * ar71xx: set priority of ath79_wdt to 1
Jul 11 19:55:01 <russell--> blogic: just tried r37251, still having startup problems with remote syslog
Jul 11 19:56:56 <russell--> added log_port and log_proto, a simple /etc/init.d/log restart fixes it.
Jul 11 20:34:09 <russell--> on boot, logread starts up, i've checked hexdump -C < /proc/$(cat /var/run/logread.1.pid)/cmdline, it looks right, netstat -a shows the udp connect "ESTABLISHED", but "logger foo" does not result in anything arriving at the remote syslog receiving host
Jul 11 20:37:24 <Wipster> hey all, I'm running 37154 on my tp-wdr3600 v1.1 (ar9344 rev2) which is a dual band router however 5Ghz doesn't come up, in the kernel log there is an invalid calibration data warning however my ART partition is intact (and backed up). Using the debugfs I can read the eeprom out from that phy and there are things in there. What do I need to do to add support for the SoC's 5Ghz radio?
Jul 11 20:57:58 <Guest92905> nbd r37252 trunk/target/ (6 files) * kernel: disable CONFIG_CLS_U32_PERF by default, it is usually not needed
Jul 11 20:58:01 <Guest92905> nbd r37253 trunk/target/ (6 files)
Jul 11 20:58:01 <Guest92905> kernel: disable CONFIG_RD_LZMA by default, it will be selected when actually needed
Jul 11 20:58:05 <Guest92905> nbd r37254 trunk/package/kernel/linux/modules/block.mk * kernel: make libsas depend on x86
Jul 11 20:58:09 <Guest92905> nbd r37255 trunk/target/ (6 files in 4 dirs)
Jul 11 20:58:10 <Guest92905> kernel: add a patch that reduces module size by removing non-essential information (reduces default rootfs size by ~5k after lzma)
Jul 11 20:58:13 <Guest92905> nbd r37256 trunk/package/kernel/linux/modules/netsupport.mk * kernel: move codel from kmod-sched-core to kmod-sched, it is rarely needed
Jul 11 21:18:08 <Guest92905> blogic r37257 trunk/package/system/procd/files/log.init * procd: make the log port default to 514 if none is specified
Jul 11 21:18:16 <Guest92905> blogic r37258 trunk/target/linux/lantiq/patches-3.8/0037-owrt-lantiq-wifi-and-ethernet-eeprom-handling.patch * lantiq: revert 36778 as it breaks irq assignment to pci devices
Jul 11 21:18:25 <Guest92905> blogic r37259 trunk/package/system/procd/Makefile * procd: update to latest git revision
Jul 11 21:18:35 <Guest92905> blogic r37260 trunk/target/ (10 files in 9 dirs) * ar71xx: add carambola2 support
Jul 11 21:40:25 <Wipster> I have found that it errors out in the ath9k_pci_fixup function as the calibration data doesn't start with the tag 0xa55a.. this might be a red herring as the pci device enables later without issue and the EEPROM gives the reg domain info
Jul 11 22:22:09 <Guest92905> hauke r37261 branches/attitude_adjustment/ (22 files in 4 dirs) * kernel: update bcma and ssb to wireless-testing/master master-2013-07-03
Jul 11 22:25:53 <Guest92905> hauke r37262 branches/attitude_adjustment/ (8 files) * AA: kernel: refresh patches
Jul 11 22:43:06 <tripolar> hi
Jul 11 22:43:29 <tripolar> i want to split my swtich into 2 seperate vlans so i did this
Jul 11 22:43:29 <tripolar> http://paste.debian.net/15626/
Jul 11 22:43:39 <tripolar> but after a restart the device is unreachable
Jul 11 22:43:50 <tripolar> do i have to change something else?
Jul 11 22:45:50 <luka> tripolar: should work, are you connected to the right port?
Jul 11 22:47:23 <tripolar> all ports are dead
Jul 11 22:47:31 <tripolar> no other changes are needed inside the file?
Jul 11 22:47:54 <luka> i dont think so
Jul 11 22:49:16 <tripolar> okay i will try it again
Jul 11 22:57:01 <tripolar> luka: it seems to work no one port is dead but ifconfig doesnt show me eth0.2 only eth0.1
Jul 11 22:57:18 <tripolar> ifconfig -a
Jul 11 23:10:08 <tripolar> blogic: do you know how can i find out whats wrong?
Jul 11 23:10:51 <luka> in lan section replace eth0.1 with eth0.2 then restart
Jul 11 23:10:57 <luka> let me know what happens then
Jul 11 23:12:02 <builder> build #336 of brcm63xx is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/336
Jul 11 23:15:55 <tripolar> luka: it seems it convigured correct
Jul 11 23:16:02 <tripolar> swconfig dev switch0 show
Jul 11 23:16:04 <tripolar> gives me
Jul 11 23:16:10 <tripolar> VLAN 1: info: VLAN 1: Ports: '0125t', members=0027, untag=0007, fid=0 fid: 0 ports: 0 1 2 5t
Jul 11 23:16:13 <tripolar> VLAN 2: info: VLAN 2: Ports: '35t', members=0028, untag=0008, fid=0 fid: 0 ports: 3 5t
Jul 11 23:17:25 <luka> how many ports are on the board?
Jul 11 23:17:33 <luka> 4, rigt?
Jul 11 23:18:49 <luka> my guess is that you were connected to port "3" and didn't plug the jack to the different one once you made the config change
Jul 11 23:24:20 <tripolar> luka: it worked to problem was i had the same name for both interfaces
Jul 11 23:27:53 <luka> good ;)
Jul 12 00:36:27 <builder> build #331 of at91 is complete: Failure [failed shell_14]  Build details are at http://buildbot.openwrt.org:8010/builders/at91/builds/331
**** ENDING LOGGING AT Fri Jul 12 02:59:58 2013