**** BEGIN LOGGING AT Fri Jul 08 02:59:58 2016 Jul 08 05:25:33 build #314 of netlogic is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/netlogic/builds/314 Jul 08 05:25:49 build #249 of adm8668 is complete: Failure [failed shell_18] Build details are at http://buildbot.openwrt.org:8010/builders/adm8668/builds/249 Jul 08 05:26:02 build #242 of mpc85xx is complete: Failure [failed shell_18] Build details are at http://buildbot.openwrt.org:8010/builders/mpc85xx/builds/242 Jul 08 05:34:36 build #269 of octeon is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/octeon/builds/269 Jul 08 06:14:55 build #282 of mxs is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/mxs/builds/282 Jul 08 06:38:35 build #319 of brcm2708 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm2708/builds/319 Jul 08 07:20:03 build #365 of brcm47xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx/builds/365 Jul 08 08:24:04 build #342 of imx6 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/imx6/builds/342 Jul 08 09:16:46 build #292 of ar7 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ar7/builds/292 Jul 08 09:21:39 build #358 of brcm63xx is complete: Failure [failed shell_18] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/358 Jul 08 10:03:50 build #270 of mcs814x is complete: Failure [failed shell_17] Build details are at http://buildbot.openwrt.org:8010/builders/mcs814x/builds/270 Jul 08 10:08:34 build #344 of ramips is complete: Failure [failed shell_17] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/344 Jul 08 12:13:10 there might be a bug - https://downloads.openwrt.org/chaos_calmer/15.05.1/brcm2708/bcm2708/packages/packages/ is missing compared w. 15.05 Jul 08 13:24:15 hmm Jul 08 13:24:34 not sure if it was removed or wasn't built at all at that time Jul 08 14:56:46 build #264 of au1000 is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/au1000/builds/264 Jul 08 15:08:39 build #268 of brcm47xx.mips74k is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm47xx.mips74k/builds/268 Jul 08 15:19:10 build #264 of ramips.rt305x is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ramips.rt305x/builds/264 Jul 08 15:37:17 https://wiki.openwrt.org/inbox/procd-init-scripts I wish i found this last week. It's the only decent instructions on how to write procd init script properly. Jul 08 15:37:46 i'm pretty sure i found a glitch with jsonfilter and ubox's ./sbin/validate_data Jul 08 15:39:37 jsonfilter adds a space to what it returns when you are after values which breaks getopt() with some programs, and /sbin/validate_data sets numeric type arguments defaults in shell to strings Jul 08 15:59:06 "I for one would hold of any further sponsorship/investment for a year or two to see if the uncertainty of the future of the project would get resolved and or to evaluate if investment in the "fork" would be better investment or look straight away into competing alternatives to openwrt and evaluate those." Jul 08 15:59:23 From https://lwn.net/Articles/686180/ the post by jow Jul 08 16:01:00 They have gone into bed with a request for 2TB for a compute cluster to do continious integration when they were only offered hosting at first by digital ocean, and now they want to have one of their main mirrors in Communist China - in the same university famous for it's reverse engineering. Looks like openwrt or at least it's fork is going the same way as huweai - banned from us government purchasing, and any embedded product based on Jul 08 16:01:01 LEDE/Openwrt too. Jul 08 16:01:55 Lede's latest plan is to buddy up with the communists at http://www.ie.tsinghua.edu.cn/eng/common.php?pid=262&ty=268 who are world famous for taking donations from any chinese company who wants to pay for any reverse engineering of a companys product. Then who are the VC people behind digital ocean? Jul 08 16:02:14 rmilecki 49398 branches/chaos_calmer/package/kernel/mac80211/patches/ (6 files) mac80211: brcmfmac: fix interfaces management Jul 08 16:03:14 http://wiki.prplfoundation.org/wiki/Procd_reference sorry this was the link to the decent howto on procd init script writing. Just QAing my work now done properly. and i'll re-do the commits to be "signed off" properly before the pull request. Jul 08 16:05:32 https://lh3.googleusercontent.com/-IMExbKyCxHg/AAAAAAAAAAI/AAAAAAAAAAA/241Z3WANpcs/photo.jpg is that jow_laptop? If it is it explains a lot about Lede's policy to block conservatives like brendan eitch (mozilla/javascript) making pull requests Jul 08 16:06:31 https://lists.openwrt.org/pipermail/openwrt-devel/2016-July/041930.html - the script linked in there is being re_done to use uci_validate_section instead. just testing now. Jul 08 16:24:51 build #352 of realview is complete: Failure [failed compile_4] Build details are at http://buildbot.openwrt.org:8010/builders/realview/builds/352 Jul 08 18:40:52 hurm Jul 08 20:21:02 https://github.com/hojuruku/openwrt-packages/blob/mini_snmpd/net/mini_snmpd/files/mini_snmpd.init any comments on the script. I'd like some feedback on how to do per instance restarts if an interface changes. Jul 08 20:29:34 i just updated it. its nearly done just for knowing how to properly do per instance interface restarts so the daemons are only reloaded if the interface they are bound to changes, whilst the other instances are not restarted Jul 08 20:45:41 unfortunately, that's getting into the even less documented corners of procd Jul 08 20:46:49 hojuruku: try looking at the init scripts for igmp proxy or something, or grep openwrt/packages for uses of service_triggers() Jul 08 20:47:03 your service_triggers seems needlessly complex at first glance Jul 08 20:47:39 procd_add_reload_trigger mini_snmpd should be enough for all of your config load stuff Jul 08 20:48:09 and you don't need the explicit reload_service() override either, that's the default. Jul 08 20:52:48 there's procd_add_network_trigger but I've never used it. Jul 08 20:54:24 karlp: i've done greps on the buildroot under the package dir for that kind of stuff Jul 08 20:54:37 karlp: i haven't seen one that does per interface restarts though that's why i'm asking Jul 08 20:54:57 who wrote procd? blogic right? Jul 08 20:55:39 no idea, I don't really care who wrote it, just about how to use it. like I said, I've never needed the network interface stuff myself. Jul 08 20:55:59 it wouldn't surprise me if things were "quirky" when you have multiple instances though Jul 08 20:57:21 then it's good enough to push, it doesn't do per_interface restarting yet, just restarts all the instances if any interface goes up and down. Jul 08 20:57:47 mini_snmpd would be good on larger systems, having many instances gets around the 4 network interface limit in it's OID Jul 08 20:58:24 i just fixed a typo. but before i do a pull request any other feedback on the script from anyone? Jul 08 20:58:35 refresh your browser if you just opened it. Jul 08 20:59:50 with github the commits as well need to have "signed-off-by" in them, because ill make a new branch to submit it again without all my draft work :P What's the best git preference to automatically do that without having to comment with you git fetch upstream? Jul 08 21:05:55 not sure you can do it automatically, you can set up a commit template that includes it though Jul 08 21:06:32 there's a format.signoff config option but it only applies to "git format-patch" for mailing cruft. Jul 08 21:06:34 git commit -m accepts \n as newlines? that will work for me. Jul 08 21:07:00 just never use commit -m, let it open $EDITOR and type a real useful valid message. Jul 08 21:07:14 i've seen perfectly good pull requests like for coova-chilli get knocked back because of that. Jul 08 21:07:39 karlp you mean like an essay? that's for the pull request. Jul 08 21:07:43 and you can just submit pull requests to get feedback, there's no real benefit to asking someone to look at it before hand, if anything it's just harder to give the feedback. Jul 08 21:08:05 no, you want it in the commit message, so it goes into the permanent record, not just some pull request commentary Jul 08 21:08:41 ok roger wilco, ill do the pull tomorrow, gotta sleep Jul 08 21:21:53 humm the rival snmpd package is treating netdev in procd as a physical device not a uci device Jul 08 21:22:19 that's the only package that uses procd_append_param netdev Jul 08 21:49:29 not sure what the param is vs the reload interface trigger Jul 09 00:46:23 build #345 of ramips is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/ramips/builds/345 Jul 09 00:56:24 build #359 of brcm63xx is complete: Success [build successful] Build details are at http://buildbot.openwrt.org:8010/builders/brcm63xx/builds/359 **** ENDING LOGGING AT Sat Jul 09 02:59:58 2016