**** BEGIN LOGGING AT Wed May 27 02:59:59 2015
May 27 06:57:58 <mrkiko> hi guys
May 27 06:58:02 <mrkiko> bye
May 27 07:38:45 * russell-- is having some troubles getting pppoe and a hurricane electric 6in4 tunnel to work at the same time, presumably due to MSS clamping issues
May 27 07:41:07 <russell--> ppp-mod-pppoe doesn't support CLAMPMSS, and rp-pppoe-client is marked broken
May 27 08:11:32 <russell--> hmm. disabling the MSS clamp doesn't fix the henet problem, so maybe unrelated
May 27 08:11:51 <russell--> i have a packet trace if anyone is interested
May 27 08:15:00 <russell--> aha
May 27 08:15:23 <russell--> ip6tables -I FORWARD -o 6in4-henet -p tcp --tcp-flags SYN,RST SYN -j TCPMSS  --clamp-mss-to-pmtu
May 27 08:15:31 <russell--> ip6tables -I FORWARD -i 6in4-henet -p tcp --tcp-flags SYN,RST SYN -j TCPMSS  --clamp-mss-to-pmtu
May 27 08:15:36 <russell--> seems to fix
May 27 08:27:19 <br101> hi! commit build: allow creating empty package feeds (15e34192d77fbb711203b92db4b439f830e5a7b7) broke package/index for me:
May 27 08:27:39 <br101> Generating index for package ./libical_1.0-1_ar71xx.ipk
May 27 08:27:39 <br101> make[1]: *** [package/index] Error 1
May 27 08:27:39 <br101> make[1]: Leaving directory `/home/br1/dev/openwrt/trunk/openwrt'
May 27 08:27:39 <br101> make: *** [package/index] Error 2
May 27 08:28:06 <br101> I guess there needs to be a exit 0 at the end of ipkg-make-index.sh
May 27 08:28:22 <jow_laptop> br101: yep
May 27 08:30:44 <OpenWrt-Commits> jow r45771 trunk/scripts/ipkg-make-index.sh * scripts: ensure that ipkg-make-index.sh exits successfully
May 27 08:34:11 <builder> build #11 of mpc83xx is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc83xx/builds/11
May 27 08:36:33 <cyrusff> russell--: mss clamping for v6?
May 27 08:36:54 <cyrusff> just set a useful mtu ;)
May 27 08:42:52 <russell--> for some reason, that doesn't work
May 27 08:43:30 <russell--> mtu is set to 1280 on the henet interface
May 27 08:43:51 <br101> jow_laptop: thanks
May 27 08:44:21 <russell--> cyrusff: that worked fine until the 6in4 tunnel was going over a pppoe interface
May 27 08:45:50 <russell--> wireshark is telling me that mss was set to 1440
May 27 08:47:28 <russell--> i guess that implies that icmp is getting eaten somehow or somehow not delivered to the client appropriately
May 27 08:48:52 <russell--> but, ping6 -s1600 git.kernel.org sees the right MTU
May 27 08:49:02 <russell--> so, /me baffled
May 27 08:51:21 <cyrusff> strange
May 27 08:51:37 * russell-- blames the NSA
May 27 08:52:46 <builder> build #11 of ep93xx is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/ep93xx/builds/11
May 27 08:59:59 <russell--> 01:58:50.289762 IP6 dremel > 2001:470:e86d:1::4: ICMP6, packet too big, mtu 1280, length 1240
May 27 09:00:20 <russell--> but, interestingly, i don't get that all the time, i wonder if there is a ratelimit on that somewhere
May 27 09:01:40 <jow_laptop> russell--: is there an itermediate openwrt router?
May 27 09:03:16 <jow_laptop> russell--: the default firewall rules limit icmp to 1000pkts/second
May 27 09:03:24 <jow_laptop> (icmpv6)
May 27 09:04:11 <russell--> jow_laptop: i'm not using the default firewall
May 27 09:04:18 <jow_laptop> ok
May 27 09:06:07 <russell--> client-device -- openwrt -- 6in4-henet_over_pppoe --- interwebs
May 27 09:08:44 <russell--> # cat /proc/sys/net/ipv6/icmp/ratelimit
May 27 09:08:46 <russell--> 1000
May 27 09:16:35 <russell--> twiddling those values randomly doesn't seem to have any effect
May 27 09:23:22 * russell-- wanders off to sleep, leaving ipv6 off for now
May 27 09:23:38 <russell--> just turned off odhcpd, actually
May 27 09:24:00 <russell--> 6in4 still up, just clients won't find it
May 27 10:05:49 <karlp> br101: yeah, was just diagnosing that package index thing myself yesterday afternoon, ran out of time and have just spent the last half hour looking at it too :| should have read more backlog first :)
May 27 10:07:41 <karlp> thanks for fixing it jow_laptop
May 27 11:23:50 <OpenWrt-Commits> jow r45772 trunk/target/ imagebuilder/Config.in imagebuilder/files/repositories.conf imagebuilder/files/Makefile imagebuilder/Makefile * IB: use online repositories
May 27 12:15:54 <builder> build #11 of mpc52xx is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc52xx/builds/11
May 27 13:01:12 <karlp> any tips on what would be causing, " * opkg_install_cmd: Cannot install package 2.1.0.
May 27 13:01:39 <karlp> with earlier, "Unknown package '2.1.0'"
May 27 13:01:51 <karlp> no clue what's triggering it all of a sudden, I guess I added a "bad" package?
May 27 13:16:43 <jow_laptop> karlp: package with underscore in the name?
May 27 13:19:36 <karlp> no idea, make clean seems to have helped, thnk its in tcpdump vs tcpdump-mini and the libpcap usb support.
May 27 13:19:49 <karlp> I've had packages with underscores in their names for a while now, never seen this before
May 27 13:21:34 <jow_laptop> was just a hunch
May 27 13:22:13 <jow_laptop> the package install code does some splitting and normalization of filenames
May 27 13:22:19 <jow_laptop> and then feeds it to xargs
May 27 13:22:39 <jow_laptop> so some version number might end op in opkg's argv if a package filename contains unexpected chars
May 27 13:23:52 <karlp> didn't know that was even something to avoid.
May 27 13:24:00 <karlp> I've got 9 packages with _ in their name
May 27 13:24:18 <karlp> using a - in boards seems to cause a problem, had to use _ there, for ar71xx at least.
May 27 13:24:38 <KanjiMonster> karlp: recent trunk? CC-rc1? selfbuilt? snapshot?
May 27 13:24:53 <karlp> KanjiMonster: this was a recent trunk, selfbuilt
May 27 13:25:09 <KanjiMonster> karlp: also during image generation or when trying to install on the device?
May 27 13:25:11 <karlp> been moving from BB to trunk to try and track down a ftdi usb issue
May 27 13:27:19 <jow_laptop> I can see three packages with version 2.1.0 here:
May 27 13:27:26 <jow_laptop> lua-bencode, expat and sysfsutils
May 27 13:27:40 <jow_laptop> neither of them looks suspicious
May 27 13:27:44 <karlp> yeah, I've got expat from avahi-nodbus,
May 27 13:27:59 <jow_laptop> whats the context of the error?
May 27 13:28:01 <karlp> clean and rebuild has fixed it, but no idea where it would have come from
May 27 13:28:05 <jow_laptop> ok
May 27 13:28:10 <jow_laptop> maybe garbage in bin/
May 27 13:28:50 <karlp> lsusb in trunk doesn't show the strings? this worked in BB? or did I install the "wrong" version of lsusb?
May 27 13:29:09 <jow_laptop> maybe the usbids.txt download failed
May 27 13:29:11 <jow_laptop> or was removed
May 27 13:29:43 <KanjiMonster> or maybe garbage in tmp/ (leftovers from BB), which caused the package defintions to be wrongly parsed
May 27 13:30:11 <karlp> no, no leftovers from BB, this is a separate tree, being a separate git repo and all that
May 27 13:32:47 <karlp> so, lsusb in trunk is _meant_ to show strings?
May 27 13:34:05 <karlp> ah, looks like my BB build had usbutils included
May 27 13:37:27 <karlp> andddd no change on 3.18, hope this is recent enough for linux-usb to listen :|
May 27 13:55:37 <OpenWrt-Commits> rmilecki r45773 trunk/tools/firmware-utils/ src/asustrx.c Makefile * firmware-utils: asustrx: new tool for TRX format with Asus tail
May 27 13:55:49 <OpenWrt-Commits> rmilecki r45774 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: build image for Asus RT-N66U
May 27 14:54:57 <OpenWrt-Commits> jow r45775 trunk/target/linux/ar7/image/Makefile * ar7: use common image prefix
May 27 14:57:43 <OpenWrt-Commits> nbd r45776 trunk/target/linux/kirkwood/patches-3.18/200-disable-tso.patch
May 27 14:57:43 <OpenWrt-Commits> kirkwood: disable TSO on the ethernet device by default, fixes data corruption issues (#19735)
May 27 14:57:53 <OpenWrt-Commits> nbd r45777 trunk/target/linux/ramips/image/Makefile * ramips: check size after padding
May 27 14:59:08 <OpenWrt-Commits> jow r45778 trunk/target/linux/arm64/image/Makefile * arm64: use common image prefix
May 27 15:05:31 <OpenWrt-Commits> jow r45779 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: use common image prefix
May 27 15:19:51 <OpenWrt-Commits> jow r45780 trunk/target/linux/brcm63xx/image/Makefile * brcm63xx: use common image prefix
May 27 15:19:58 <OpenWrt-Commits> jow r45781 trunk/target/linux/cns3xxx/image/Makefile * cns3xxx: use common image prefix
May 27 15:20:06 <OpenWrt-Commits> jow r45782 trunk/target/linux/gemini/image/Makefile * gemini: use common image prefix
May 27 15:21:34 <OpenWrt-Commits> nbd r45783 trunk/target/linux/generic/patches-3.18/ 645-bridge_multicast_to_unicast.patch 051-01-bridge-fix-parsing-of-MLDv2-reports.patch 051-02-bridge-allow-setting-hash_max-multicast_router-if-in.patch * kernel: bridge: backport two snooping related patches
May 27 15:21:52 <OpenWrt-Commits> nbd r45784 trunk/scripts/feeds * scripts/feeds: fix typo in copied code
May 27 15:35:49 <OpenWrt-Commits> jow r45785 trunk/target/linux/ixp4xx/image/Makefile * ixp4xx: use common image prefix
May 27 15:36:00 <OpenWrt-Commits> jow r45786 trunk/target/linux/omap/image/Makefile * omap: use common image prefix
May 27 15:36:07 <OpenWrt-Commits> jow r45787 trunk/target/linux/realview/image/Makefile * realview: use common image prefix
May 27 15:36:20 <OpenWrt-Commits> jow r45788 trunk/target/linux/sunxi/image/Makefile * sunxi: use common image prefix
May 27 15:38:34 <OpenWrt-Commits> rmilecki r45789 trunk/target/linux/brcm47xx/image/Makefile * brcm47xx: build images for Asus devices
May 27 16:08:12 <karlp> nbd: updated the feeds -p, but it's made a new patchwork bit, is there a way to update the existing patchwork ticket? or is it correct to have a new one for each version of a patch?
May 27 16:10:14 <KevinDB> karlp: I think that's normal...or it did it for me too which probably means I did something wrong...again :-)  So I went into patchwork and removed the unwanted one.
May 27 16:11:19 <KanjiMonster> karlp: yes, expected. Assuming you have a patchwork acocunt, you should then mark your older version as superseded (if the reviewer didn't set it to "changes requested" already)
May 27 16:11:44 <karlp> right, I've marked the old one superseded now, theyd already marked it as changes requested.
May 27 16:21:35 <OpenWrt-Commits> rmilecki r45790 trunk/target/linux/bcm53xx/image/Makefile * bcm53xx: build images for Asus devices
May 27 17:59:40 <OpenWrt-Commits> rmilecki r45791 trunk/tools/firmware-utils/src/asustrx.c * firmware-utils: asustrx: support setting firmware version
May 27 19:04:53 <OpenWrt-Commits> kaloz r45792 trunk/target/linux/ mvebu/base-files/etc/init.d/u-boot_env mvebu/base-files/lib/upgrade/linksys.sh mvebu/base-files/etc/init.d * mvebu: make flashing more safe
May 27 19:54:59 <builder> build #13 of cobalt is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/cobalt/builds/13
May 27 19:55:01 <builder> build #13 of pxa is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/pxa/builds/13
May 27 19:55:05 <builder> build #13 of orion is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/13
May 27 19:58:14 <builder> build #13 of ppc40x is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/ppc40x/builds/13
May 27 20:03:34 <OpenWrt-Commits> blogic r45793 trunk/target/linux/ (6 files in 6 dirs) * ramips: diable ill_acc driver by default
May 27 20:06:33 <OpenWrt-Commits> blogic r45794 trunk/target/linux/ipq806x/config-3.18 * ipq806x: add missing symbol
May 27 20:07:35 <OpenWrt-Commits> blogic r45795 trunk/target/linux/ramips/patches-3.18/999-disable_illacc.patch * ramips: dont set illacc to y by default
May 27 20:08:34 <builder> build #13 of cns21xx is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/cns21xx/builds/13
May 27 21:27:48 <builder> build #12 of mpc85xx is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/12
May 27 21:34:37 <builder> build #13 of ipq806x is complete: Failure [failed compile_4]  Build details are at http://buildbot.openwrt.org:8010/builders/ipq806x/builds/13
May 27 22:08:50 <builder> build #11 of adm8668 is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/11
May 27 22:33:45 <builder> build #12 of avr32 is complete: Failure [failed shell_10]  Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/12
May 27 23:32:04 <builder> build #11 of mcs814x is complete: Success [build successful]  Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/11
**** ENDING LOGGING AT Thu May 28 02:59:58 2015