**** BEGIN LOGGING AT Wed Aug 13 02:59:58 2014 Aug 13 04:16:28 build #241 of mvebu is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/mvebu/builds/241 Aug 13 04:34:34 Swalker: not yet. Ok, thanks. Aug 13 08:20:32 KanjiMonster: can you help me figure out the ramips led problem? Aug 13 08:22:19 i just flashed a vanilla r42153 on it Aug 13 08:36:54 build #611 of ixp4xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ixp4xx/builds/611 Aug 13 09:06:11 hmm. this might be relevant: Documentation/devicetree/bindings/gpio/gpio.txt Aug 13 10:11:43 build #587 of au1000 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/587 Aug 13 11:11:37 build #551 of adm5120 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/adm5120/builds/551 Aug 13 11:11:41 Mazilo, freeswitch is still broken :( Aug 13 11:12:33 cyrusff, are there any logs from building rc3? Aug 13 11:12:45 slachta: for freeswitch? Aug 13 11:12:58 cyrusff, for the whole telephony repository Aug 13 11:13:13 or at least something like buildbot style logs Aug 13 11:19:28 forgot to try BB repository, I am always building the trunk Aug 13 11:33:15 seemed the builder somehow are the logs Aug 13 11:33:25 but at least asterisk is there in RC3 again Aug 13 11:34:27 cyrusff, let me know about any breakage in telephony, I will resolve all issues asap (excep the freeswitch I am not maintaining) Aug 13 11:34:34 sure Aug 13 11:34:42 http://downloads.openwrt.org/barrier_breaker/14.07-rc3/ar71xx/generic/packages/ Aug 13 11:34:46 anything missing there? Aug 13 11:34:48 aside freeswitch? Aug 13 11:36:58 but regarding freeswitch Aug 13 11:37:17 according to the snapshot buildbots its still broken because it cannot find a c-compiler Aug 13 11:38:25 http://buildbot.openwrt.org:8010/broken_packages/adm5120/freeswitch/compile.txt Aug 13 11:38:26 etc. Aug 13 11:39:51 I know about that. Aug 13 11:39:56 Mazilo, ping Aug 13 11:40:24 slachta: pong Aug 13 11:41:13 Mazilo: freeswitch still doesn't compile with the BB builder Aug 13 11:41:18 I've the last change you made to freeswitch, but the compilation still fails (c compiler cannot be found - libs/srtp issue) Aug 13 11:41:22 it cannot find a c-compiler in the configure stage Aug 13 11:41:25 I've tried* Aug 13 11:43:21 slachta: On my compilation, I don't see that problem. However, when I used the .config file from trine, the compilation crashes as you described. Aug 13 11:44:05 slachta: So far, the difference between my .config file and trine's is due to CONFIG_BINUTILS_VERSION_LINARO=y. Aug 13 11:44:17 Mazilo, I used the default config with all subpackages enabled Aug 13 11:45:57 slachta: Right now, I am rebuilding from scratch to see if setting CONFIG_BINUTILS_VERSION_LINARO=y will fix the issue. If so, I will change the SRTP settings based on the CONFIG_BINUTILS_VERSION_LINARO=y setting. Aug 13 11:46:40 cyrusff, there is a half of asterisk (both versions - 1.8 and 11.x) packages, seems that compilation failed. Can you take a look at the compilation logs if it is caused by missing mysql headers? Aug 13 11:47:19 Mazilo: make sure you enable ccache in your advanced settings in menuconfig Aug 13 11:47:29 slachta: as i said i don't have logs unfortunately Aug 13 11:47:36 the builder ate them for some reason Aug 13 11:47:41 hope to have some soon Aug 13 11:48:07 Mazilo: that's more of a workaround than a fix, can you try to find out why it thinks it can't create exectuables? (look at the configure.log, ti should have something in there Aug 13 11:49:24 cyrusff: In my original .config file, I don't use ccache and libs/srtp compiles OK. Aug 13 11:49:44 that can be, then please try to fix it with ccache enabled Aug 13 11:49:51 we cannot build without ccache Aug 13 11:49:56 or well we don't want to Aug 13 11:50:26 because that would increase build time unnecessarily Aug 13 11:52:46 Right now, I am rebuilding from scratch without ccache. I will try with ccache after the rebuild stops and/or finishes. Aug 13 12:33:15 the branch name is after semicolon in feeds.conf, right? Aug 13 12:33:29 yes Aug 13 12:41:05 I sent a patch to mailing list to change the telephony branch Aug 13 12:41:16 to for-14.07 from master Aug 13 13:46:31 <_trine> there is a problem with trunk Aug 13 13:46:50 <_trine> specifically with /etc/config/system Aug 13 13:48:48 <_trine> even though I have trunk/files/etc/config/system the system file gets generated with both the content I have in my own trunk/files/etc/config/system in addtion to what it puts in itself so in effect it doubles the configuration Aug 13 14:04:07 cyrus r42159 branches/barrier_breaker/feeds.conf.default * BB: feeds: switch telephony feed to for-14.07 branch Aug 13 14:32:38 _trine: that's possibly because there's scripts that are adding things to /e/c/system, not just a template file that yours would then overwrite Aug 13 14:35:27 <_trine> hi karlp well it is adding to my file and not overwriting it Aug 13 14:36:11 _trine: what exactly is it adding? Aug 13 14:36:12 <_trine> I would have thought this should not happen if you have your own file located in /files Aug 13 14:36:26 _trine: grep system /rom/etc/uci-defaults/* Aug 13 14:37:05 <_trine> KanjiMonster, it is adding what the default file would be to the one I have in /files Aug 13 14:38:23 _trine: do you see the same in build_dir/target-*/root-*/etc/config/system ? Aug 13 14:38:55 <_trine> sorry for delays in responding ,, I have the nano on a different computer to this one Aug 13 14:40:11 <_trine> KanjiMonster, no its ok there Aug 13 14:42:56 <_trine> jow_laptop, there isnt a system file in /rom/etc/uci-defaults/* Aug 13 14:43:16 no, but there might be scripts there that do things like "uci set system.blah=wop" Aug 13 14:43:28 which fits with you saying that things are getting _added_ Aug 13 14:43:34 <_trine> yes Aug 13 14:44:26 on ar71xx there is (/rom)/etc/uci-defaults/04_led_migration at least which adds stuff to /e/c/system Aug 13 14:44:38 does that fit your situation? Aug 13 14:45:07 <_trine> I thing it is the file /rom/etc/uci-defaults/leds Aug 13 14:45:15 <_trine> which is causing the problem Aug 13 14:45:19 what platform? Aug 13 14:45:26 <_trine> ar71xx Aug 13 14:45:30 <_trine> nanostation m2 Aug 13 14:46:02 yes, I see Aug 13 14:46:02 <_trine> think* Aug 13 14:46:45 quick workaround: add an empty files/etc/uci-defaults/02_leds Aug 13 14:47:16 or maybe not empty but at least containing "exit 0" to signal success so its getting deleted Aug 13 14:49:01 <_trine> well I can edit the file on the nano as a quick fix Aug 13 14:49:16 yeah but the above would be a permanent solution for your custom builds Aug 13 14:49:35 simply shadow the offending uci-defaults script with a varaient in files/ that does nothing Aug 13 14:49:41 *variant Aug 13 14:50:22 can't think of something cleaner... in the long run those uci-defaults scripts need to check whether they're operating on the original Aug 13 14:50:32 but that requires some sort of checksumming Aug 13 14:51:17 maybe by comparing the current md5sum with the one from the opkg conffiles metadata (which might have been stripped away at build time) Aug 13 14:51:21 <_trine> http://paste.debian.net/115380/ Aug 13 14:51:29 <_trine> thats what it has at the moment Aug 13 14:51:34 I thought I'd heard earlier that putting conflicting files in your own files/ directory wouldn't always work, or is that for packages? where package provided files have no guarantee of being installed at any specific order? Aug 13 14:51:49 <_trine> you can see what I want and the added extra afterwards Aug 13 14:51:52 karlp: files/ always comes last Aug 13 14:52:10 ok, good to know Aug 13 14:52:13 karlp: problem is with firstboot scripts that do inplace amendments in configs, e.g. for hw detection Aug 13 14:52:18 jow_laptop: although I wonder how the 02_leds could cause it, the add_led_ helpers check first if the section they want to add already exist and if so, don't do anything Aug 13 14:52:24 yeah, I have this too, Aug 13 14:53:44 KanjiMonster: true Aug 13 14:53:56 _trine: can you pastebin the faulty system file with duplicate contents? Aug 13 14:55:40 <_trine> jow_laptop, do you mean /rom/etc/uci-defaults/leds Aug 13 14:55:49 no /etc/config/system Aug 13 14:55:56 <_trine> ok will do Aug 13 14:56:32 <_trine> I will have to reset the nano 1 minute Aug 13 14:57:10 cyrus r42160 trunk/package/network/services/odhcpd/Makefile * odhcpd: improve DHCPv6-PD detection Aug 13 14:57:14 cyrus r42161 trunk/package/network/config/netifd/Makefile * netifd: assign ipv6-prefixes with length <64 with /64 on-link routes Aug 13 15:07:22 <_trine> jow_laptop, http://paste.debian.net/115387/ Aug 13 15:08:13 <_trine> and for some reason I keep being disconnected in putty with this message Aug 13 15:08:22 <_trine> Disconnected: Server protocol/violation: unexpected SSH2_MSG_UNIMPLEMENTED packet Aug 13 15:18:03 _trine: yeah, theres a bug for this ssh thing yet Aug 13 15:18:22 <_trine> ok so long as you are aware Aug 13 15:20:54 _trine: the issue is that the script uses the name of the leds for identification, you named them differently so the 02_leds thinks they don't exist yet Aug 13 15:21:11 <_trine> ok Aug 13 15:22:03 the 02_leds expects the uppercased names, so if you use them you should have no duplicate entries anymore Aug 13 15:22:52 <_trine> KanjiMonster, I will give that a try thanks Aug 13 17:41:14 build #655 of ppc44x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/ppc44x/builds/655 Aug 13 18:07:08 <_trine> KanjiMonster, is this how you meant Aug 13 18:07:12 <_trine> http://paste.debian.net/115441/ Aug 13 18:07:42 <_trine> because it still appends the original file to my new one Aug 13 18:07:49 _trine: no, I meant using 'RSSILOW', 'RSSIMEDIUMLOW' etc for name Aug 13 18:08:02 the sysfs entries probably need to stay lowercase Aug 13 18:08:17 <_trine> oh I see I thought you just meant to do it all in uppercase Aug 13 18:08:29 <_trine> I will give it another try Aug 13 18:08:57 <_trine> so you are saying I have to use those specif names as well as keeping the same case Aug 13 18:17:26 <_trine> KanjiMonster, http://paste.debian.net/115442/ Aug 13 18:17:31 <_trine> hows that Aug 13 18:18:04 <_trine> I left a '3' in took it out now Aug 13 18:18:29 _trine: the name should now prevent the double addition, but the leds won't likely work as the sysfs values must likely be lowercase (to match the actual file names) Aug 13 18:19:12 <_trine> OK I will alter those as well Aug 13 18:20:38 <_trine> http://paste.debian.net/115443/ Aug 13 18:20:43 <_trine> hows that Aug 13 18:21:12 <_trine> I will compile that and give it a try Aug 13 18:22:44 <_trine> ah you said lower case for the sysfs Aug 13 18:25:36 yes, they still need to be 'ubnt::' Aug 13 18:26:14 <_trine> this should be it Aug 13 18:26:17 <_trine> http://paste.debian.net/115444/ Aug 13 18:28:48 yes, this should work Aug 13 18:29:13 <_trine> I will let you know Aug 13 18:30:32 <_trine> really there should be check made to see if /files/etc/config/system exists and if it does then use that and nothing else Aug 13 18:32:41 <_trine> this bug in SSH2 is bugging me Aug 13 18:50:52 build #261 of imx6 is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/imx6/builds/261 Aug 13 19:07:13 <_trine> KanjiMonster, http://paste.debian.net/115456/ same Aug 13 19:08:37 _trine: then try taking over the secion names (i.e. 'led_rssimediumlow') Aug 13 19:17:25 build #666 of uml is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/uml/builds/666 Aug 13 19:26:15 <_trine> KanjiMonster, http://paste.debian.net/115461/ I am try this now Aug 13 19:26:20 <_trine> trying Aug 13 19:26:40 if this doesn't help then I'm out of ideas Aug 13 19:26:46 wait Aug 13 19:26:57 _trine: line 38 is wrong Aug 13 19:27:10 <_trine> ah yes Aug 13 19:27:12 <_trine> ' Aug 13 19:27:17 and a led Aug 13 19:27:37 config led_rssimediumlow' => config led 'led_rssimediumlow' Aug 13 19:27:37 <_trine> yeah Aug 13 19:27:58 <_trine> corrected now Aug 13 19:28:02 <_trine> I will compile Aug 13 19:29:04 <_trine> its annoying if you forget to remove system~ because that get included in the image Aug 13 19:29:06 <_trine> :P Aug 13 19:29:31 <_trine> another check needed Aug 13 19:29:34 <_trine> :) Aug 13 19:29:43 <_trine> for old people Aug 13 19:30:29 <_trine> KanjiMonster, I have another interesting bug Aug 13 19:30:36 <_trine> unrelated to all this Aug 13 19:30:41 <_trine> but interesting Aug 13 19:31:01 <_trine> however I think it might be hard to track down Aug 13 19:31:11 <_trine> do you want to hear about it? Aug 13 19:33:57 _trine: feel free to talk about it, but don't expect that I will do anything about it (or even accept it as a bug ;p) Aug 13 19:34:12 otoh sometimes strange things have very simple explanations Aug 13 19:34:22 <_trine> I dont mind either way Aug 13 19:34:38 <_trine> but it is one of those hard to understand bugs Aug 13 19:34:42 <_trine> so here goes Aug 13 19:35:01 <_trine> in this instance i am talking about my nanotion m2 Aug 13 19:35:30 <_trine> for all intents and purposes it seems to work normally except for one thing Aug 13 19:36:17 <_trine> to start this story I have 2 firmware builds one from November 2013 and todays trunk Aug 13 19:37:06 <_trine> there is a particular AP not far away that I used to be able to connect with a good fast wifi connection Aug 13 19:38:27 <_trine> now with todays trunk it will connect quickly but it will not pass data hardly at all from the internet making it impossible because of the exceptional slowness to browse the net Aug 13 19:38:50 <_trine> however if i do udhcpc -i eth0 it answers instantly Aug 13 19:39:05 <_trine> showing the connection is good Aug 13 19:39:13 <_trine> that is case 1 Aug 13 19:39:16 <_trine> case 2 Aug 13 19:39:46 <_trine> I reflash my nanostation with the firmware I compiled in November 2013 Aug 13 19:40:00 <_trine> I keep the same exact config Aug 13 19:40:16 * russell-- notices KanjiMonster talking about leds! Aug 13 19:40:27 <_trine> now I can connect to this AP and have a very good connection to the net in every way Aug 13 19:41:13 <_trine> going back to case 1 and trunk with other AP's it is connected to it seems to behave normally Aug 13 19:41:21 <_trine> so now I dont know why Aug 13 19:41:47 ah, so it is a wifi problem Aug 13 19:41:48 <_trine> trunk is not allowing me to browse like the November 2013 firmware Aug 13 19:41:50 well no idea ;p Aug 13 19:42:16 <_trine> I would not say it was a wifi problem for sure Aug 13 19:42:28 * russell-- still trying to figure out why the ramips awm003 dts LED definitions don't result in sysfs LEDs Aug 13 19:42:34 <_trine> because it only seems to happen for this 1 particular AP Aug 13 19:45:57 _trine: what wifi hardware are you using to connect Aug 13 19:45:57 ? Aug 13 19:46:06 <_trine> nanostation Aug 13 19:46:08 <_trine> m2 Aug 13 19:46:20 <_trine> ar71xx Aug 13 19:46:37 so nanostation <-> nanostation? Aug 13 19:47:15 <_trine> no the other AP is unknown :~@ Aug 13 19:47:45 <_trine> I could nmap it Aug 13 19:48:38 so it isn't even yours? ;p Aug 13 19:48:48 <_trine> its not mine Aug 13 19:49:03 <_trine> does that count? Aug 13 19:49:10 * _trine laughs Aug 13 19:49:32 <_trine> but i dont think it is relevant to the problem Aug 13 19:50:49 it's relevant insofar as it's not under your control and haven't/can't reflash it at will Aug 13 19:51:22 <_trine> exactly Aug 13 19:51:39 <_trine> I guess I might be able to reflash it Aug 13 19:51:46 <_trine> but i am not Aug 13 19:51:48 what is its bssid (or the first three octets?) Aug 13 19:52:02 <_trine> its not your russel Aug 13 19:52:35 <_trine> I am in the middle of flashing my nano Aug 13 19:53:29 nov 2013 is only 9 months ago, you could try bisecting Aug 13 19:54:57 <_trine> KanjiMonster, the led problem is now ok Aug 13 19:55:04 great Aug 13 20:00:13 <_trine> the only thing that crosses my mind regarding the odd connection problem it was about that time I stopped compiling into my code a fix for the regulatory stuff to give me channel 14 etc but I cannot imagine that to be relevant although it is on channel 1 2.412GHz Aug 13 20:01:01 _trine: D-Link Aug 13 20:02:02 <_trine> I need a d-link mac address so I can keep on using dyndns Aug 13 20:02:41 <_trine> they are going to kick people off in the next couple of months unless they pay or have a d-link router Aug 13 20:03:41 dyndns should never see your macaddr Aug 13 20:04:48 wow, they really turned that into a pay model. Aug 13 20:05:05 <_trine> remember they stopped providing a free service a few months back then someone noticed if you re registered as though you had a d-link router you could get back on board again still free Aug 13 20:06:02 <_trine> now they have sent everyone a new Email message that we all have to provide our mac addresses otherwise it's the be boot again Aug 13 20:06:14 <_trine> now they have sent everyone a new Email message that we all have to provide our mac addresses otherwise it's the big boot again Aug 13 20:06:35 <_trine> sad but true Aug 13 20:09:30 <_trine> nearly as sad as the fact that our Chief of police has just been charged with a serious criminal offence and he's still working in the job :P Aug 13 20:09:48 well,it's not exactly a monopoly anymore anyway, unless you were foolish/unlucky enough to have been relying on an actuall blah.dyndns.org domain name. Aug 13 20:10:04 people should _always_ have their own domains, I'm always surprised at this. Aug 13 20:10:43 well, our minister of the interior refused to even step aside while the police under her charge investigated her and her office for leaks Aug 13 20:10:57 <_trine> karlp, I have a fixed IP so I am ok but I did like the dns names i had which will no doubt be already taken elsewhere Aug 13 20:10:57 corruption is a global resource :) Aug 13 20:11:20 so get your own damn name :) Aug 13 20:11:54 <_trine> karlp, its so bad around here I keep a website for all the LOCAL cops who have been sent to prison Aug 13 20:11:56 <_trine> lol Aug 13 20:12:26 <_trine> its worse than the Italian mafia Aug 13 20:12:39 where's here if I amy ask? Aug 13 20:12:48 <_trine> Manchester UK Aug 13 20:13:32 <_trine> http://www.manchestereveningnews.co.uk/news/greater-manchester-news/second-mp-joins-calls-chief-7605179 Aug 13 20:13:56 * russell-- notices this discussion has no bearing whatsoever on my missing /sys/class/leds/$foo Aug 13 20:18:29 <_trine> by the way this is the same chief og police Sir Peter Fahy in a dress http://www.chiefinquisitor.webspace.virginmedia.com/fahy-drag.jpg Aug 13 20:18:35 <_trine> of* Aug 13 20:18:57 now, there's nothing wrong with a good frock, don't get distracted. Aug 13 20:19:00 <_trine> looks quite nice on a dark night Aug 13 20:19:08 <_trine> hehe Aug 13 20:20:11 <_trine> maybe he will get charged for turd burgling as well Aug 13 20:24:27 somebody working on the wiki right now? i get "DokuWiki Setup Error" from http://wiki.openwrt.org/ :( Aug 13 20:25:19 <_trine> ditto Aug 13 20:28:40 i am supposing that the /sys/class/leds/$foo should show up despite there being no uci led configuration Aug 13 20:50:02 jogo r42162 trunk/package/network/services/dropbear/patches/001-dont_send_SSH_MSG_IMPLEMENTED.patch * dropbear: fix keepalive with putty Aug 13 20:51:11 <_trine> thanks jogo Aug 13 20:51:55 _trine: you're welcome Aug 13 20:52:51 <_trine> now i have to flash again Aug 13 20:54:04 <_trine> KanjiMonster, you have had a successful day today Aug 13 22:43:30 wiki back Aug 13 22:43:31 can anyone change the topic back, please? Aug 13 22:43:38 (in #openwrt) Aug 13 22:51:35 jow_laptop: bit of a late reply, but cheers, that clears the difference up. Aug 13 23:27:11 build #624 of sibyte is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/sibyte/builds/624 Aug 14 00:15:53 build #636 of avr32 is complete: Failure [failed compile_5] Build details are at http://buildbot.openwrt.org:8010/builders/avr32/builds/636 Aug 14 01:24:20 build #62 of brcm47xx.legacy is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx.legacy/builds/62 Aug 14 01:31:03 build #701 of orion is complete: Failure [failed compile_8] Build details are at http://buildbot.openwrt.org:8010/builders/orion/builds/701 Aug 14 02:01:56 build #560 of iop32x is complete: Failure [failed shell_10] Build details are at http://buildbot.openwrt.org:8010/builders/iop32x/builds/560 **** ENDING LOGGING AT Thu Aug 14 02:59:59 2014